Perceptual Generative Adversarial Networks for Small Object Detection
Open accessAbstract
Detecting small objects is notoriously challenging due to their low resolution and noisy representation. Existing object detection pipelines usually detect small objects through learning representations of all the objects at multiple scales. However, the performance gain of such ad hoc architectures is usually limited to pay off the computational cost. In this work, we address the small object detection problem by developing a single architecture that internally lifts representations of small objects to super-resolved ones, achieving similar characteristics as large objects and thus more discriminative for detection. For this purpose, we propose a new Perceptual Generative Adversarial Network (Perceptual GAN) model that improves small object detection through narrowing representation difference of small objects from the large ones. Specifically, its generator learns to transfer perceived poor representations of the small objects to super-resolved ones that are similar enough to real large objects to fool a competing discriminator. Meanwhile its discriminator competes with the generator to identify the generated representation and imposes an additional perceptual requirement - generated representations of small objects must be beneficial for detection purpose - on the generator. Extensive evaluations on the challenging Tsinghua-Tencent 100K [45] and the Caltech [9] benchmark well demonstrate the superiority of Perceptual GAN in detecting small objects, including traffic signs and pedestrians, over well-established state-of-the-arts.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
For small object detection, GAN can be used to enhance the features of small objects by narrowing the representations between small and large ones.
“For small object detection, GAN can be used to enhance the features of small objects by narrowing the representations between small and large ones [198, 199].”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [198]
Many recent pedestrian detection methods are extensions of Fast/Faster R-CNN and profit from the basic concepts of these methods.
“Many recent pedestrian detection methods [62], [63], [71], [55], [70], [57], [56] are extensions of Fast/Faster R-CNN and profit from the basic concepts of these methods.”
From EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes (2019) · cites this paper as [62]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1× · see citing works
“Following [16], we perform down-sampling directly by convolutional layers with a stride of 2.”
MethodVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 74,248 · cited 1× · see citing works
“For pedestrian detection, we use the pretrained VGG-16 model [35] as adopted in [41].”
MethodFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812 · cited 5× · see citing works
“For the Tsinghua-Tencent 100K [45] benchmark, we train a Region Proposal Network (RPN) as proposed in [32] to generate object proposals on the training and testing images.”
BackgroundMethodInfluentialFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 56,035 · cited 5× · see citing works
“For the Tsinghua-Tencent 100K [45] benchmark, we train a Region Proposal Network (RPN) as proposed in [32] to generate object proposals on the training and testing images.”
BackgroundMethodInfluentialHistograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2× · see citing works
“We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
InfluentialRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684 · cited 1× · see citing works
“Following the parameterization scheme in [12], rk specifies a scale-invariant translation and log-space height/width shift relative to an object proposal.”
Method(2016) · Cited by 31,590 · see citing works
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105 · see citing works
Labaca-Castro, R. (2023) · Springer eBooks · doi:10.1007/978-3-658-40442-0_9 · Cited by 21,100 · see citing works
The pascal visual object classes (VOC) challenge
Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-009-0275-4 · Cited by 19,599 · see citing works
GAN（Generative Adversarial Nets）
淳司, 柴. (2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 19,268 · see citing works
(2010) · Cited by 14,134 · see citing works
Robust Real-Time Face Detection
Viola, P., & Jones, M. J. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000013087.49260.fb · Cited by 14,097 · see citing works
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Ledig, C., Theis, L., Huszar, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., Wang, Z., et al. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.19 · Cited by 12,696 · cited 1× · see citing works
“GANs were also applied to image superresolution in [21].”
MethodUnsupervised representation learning with deep convolutional generative adversarial networks
Radford, A., Metz, L., & Chintala, S. (2015) · arXiv · doi:10.48550/arxiv.1511.06434 · Cited by 12,235 · cited 1× · see citing works
“The idea of using GANs for unsupervised representation learning was described in [31].”
MethodCaffe: Convolutional Architecture for Fast Feature Embedding
Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., & Darrell, T. (2014) · Proceedings of the 22nd ACM international conference on Multimedia · doi:10.1145/2647868.2654889 · Cited by 11,077 · cited 1× · see citing works
“The implementation is based on the publicly available Fast R-CNN framework [11] built on the Caffe platform [17].”
MethodReturn of the Devil in the Details: Delving Deep into Convolutional Nets
Chatfield, K., Simonyan, K., Vedaldi, A., & Zisserman, A. (2014) · Proceedings of the British Machine Vision Conference 2014 · doi:10.5244/c.28.6 · Cited by 3,335 · cited 1× · see citing works
“For traffic sign detection, we use the pretrained VGGCNN-M-1024 model [3] as adopted in [24] to initialize our network.”
MethodPedestrian Detection: An Evaluation of the State of the Art
Dollár, P., Wojek, C., Schiele, B., & Perona, P. (2011) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2011.155 · Cited by 3,191 · cited 8× · see citing works
“For the Caltech benchmark [9], we utilize the ACF pedestrian detector [7] trained on the Caltech training set for object proposals generation.”
BackgroundMethodInfluentialDeep Generative Image Models using a Laplacian Pyramid of Adversarial NetworksPreprint
Denton, E., Chintala, S., Szlam, A., & Fergus, R. (2015) · Neural Information Processing Systems · doi:10.48550/arxiv.1506.05751 · Cited by 2,360 · cited 1× · see citing works
“[6] adopted GANs for the application of image generation.”
MethodFast Feature Pyramids for Object Detection
Dollár, P., Appel, R., Belongie, S., & Perona, P. (2014) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2014.2300479 · Cited by 1,941 · cited 3× · see citing works
“proposed Integral Channel Features (ICF) [8] and Aggregated Channel Features (ACF) [7], which are among the most popular hand-crafted features for constructing pedestrian detectors.”
BackgroundMethodA convolutional neural network cascade for face detection
Li, H., Lin, Z., Shen, X., Brandt, J., & Hua, G. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299170 · Cited by 1,390 · cited 2× · see citing works
“Some efforts [4, 25, 18, 39, 23, 1] have been devoted to addressing small object detection problems.”
BackgroundInside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks
Bell, S., Zitnick, C. L., Bala, K., & Girshick, R. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.314 · Cited by 1,344 · cited 3× · see citing works
“As shown in Table 3, “Skip Pooling” indicates the model trained by combining low-level features through skip pooling as proposed in [1].”
BackgroundMethodDollar, P., Tu, Z., Perona, P., & Belongie, S. (2009) · Procedings of the British Machine Vision Conference 2009 · doi:10.5244/c.23.91 · Cited by 1,237 · cited 2× · see citing works
“proposed Integral Channel Features (ICF) [8] and Aggregated Channel Features (ACF) [7], which are among the most popular hand-crafted features for constructing pedestrian detectors.”
BackgroundDeep multi-scale video prediction beyond mean square errorPreprint
Mathieu, M., Couprie, C., & LeCun, Y. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.05440 · Cited by 1,182 · see citing works
Traffic-Sign Detection and Classification in the Wild
Zhu, Z., Liang, D., Zhang, S., Huang, X., Li, B., & Hu, S. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.232 · Cited by 1,062 · cited 10× · see citing works
“[45] in terms of average recall and accuracy: 89% and 84% vs 87% and 82%, 96% and 91% vs 94% and 91%, 89% and 91% vs 88% and 91% on three subsets of different object sizes.”
BackgroundMethodInfluentialIs Faster R-CNN Doing Well for Pedestrian Detection?
Zhang, L., Lin, L., Liang, X., & He, K. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46475-6_28 · Cited by 852 · cited 2× · see citing works
“Recently, deep learning methods have greatly boosted the performance of pedestrian detection [29, 33, 28, 36, 41].”
BackgroundMethodSelf-citationDeep Relative Distance Learning: Tell the Difference between Similar Vehicles
Liu, H., Tian, Y., Wang, Y., Pang, L., & Huang, T. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.238 · Cited by 806 · cited 1× · see citing works
“For traffic sign detection, we use the pretrained VGGCNN-M-1024 model [3] as adopted in [24] to initialize our network.”
MethodTraffic sign recognition with multi-scale Convolutional Networks
Sermanet, P., & LeCun, Y. (2011) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2011.6033589 · Cited by 771 · cited 2× · see citing works
“[34] proposed to feed multi-stage features to the classifier using connections that skip layers to boost traffic sign recognition.”
BackgroundCombining Markov Random Fields and Convolutional Neural Networks for Image Synthesis
Li, C., & Wand, M. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.272 · Cited by 758 · cited 1× · see citing works
“In [22] and [40], GANs were employed to learn a mapping from one manifold to another for style transfer and inpainting, respectively.”
MethodPedestrian Detection with Unsupervised Multi-stage Feature Learning
Sermanet, P., Kavukcuoglu, K., Chintala, S., & LeCun, Y. (2013) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2013.465 · Cited by 753 · cited 2× · see citing works
“Recently, deep learning methods have greatly boosted the performance of pedestrian detection [29, 33, 28, 36, 41].”
BackgroundMethodJoint Deep Learning for Pedestrian Detection
Ouyang, W., & Wang, X. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.257 · Cited by 665 · cited 2× · see citing works
“[29] proposed a deformation hidden layer for CNN to model mixture poses information, which can further benefit the pedestrian detection task.”
Background(2015) · Cited by 656 · see citing works
Yang, F., Choi, W., & Lin, Y. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.234 · Cited by 608 · cited 2× · see citing works
“Some efforts [4, 25, 18, 39, 23, 1] have been devoted to addressing small object detection problems.”
BackgroundTen Years of Pedestrian Detection, What Have We Learned?
Benenson, R., Omran, M., Hosang, J., & Schiele, B. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16181-5_47 · Cited by 599 · cited 2× · see citing works
“We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
MethodInfluentialHow Far are We from Solving Pedestrian Detection?
Zhang, S., Benenson, R., Omran, M., Hosang, J., & Schiele, B. (2016) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2016.141 · Cited by 473 · cited 1× · see citing works
“The evaluation metric is log-average Miss Rate on False Positive Per Image (FPPI) in [10−2, 10] (denoted as MR following [42]).”
Pedestrian detection aided by deep learning semantic tasks
Tian, Y., Luo, P., Wang, X., & Tang, X. (2015) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2015.7299143 · Cited by 416 · cited 4× · see citing works
“We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
BackgroundMethodInfluentialFiltered channel features for pedestrian detection
Zhang, S., Benenson, R., & Schiele, B. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298784 · Cited by 395 · cited 2× · see citing works
“We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
InfluentialLocal Decorrelation for Improved Pedestrian Detection
Nam, W., Stradvision Inc, & Han, J. H. (2015) · Cited by 336 · see citing works
Learning Complexity-Aware Cascades for Deep Pedestrian Detection
Cai, Z., Saberian, M., & Vasconcelos, N. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.384 · Cited by 326 · cited 3× · see citing works
“We use dense sampling of the training data (every 4th frame) as adopted in [44, 27].”
MethodInfluentialTraffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks
Jin, J., Fu, K., & Zhang, C. (2014) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2014.2308281 · Cited by 310 · cited 2× · see citing works
“[19] proposed to train the CNN with hingle loss, which provides better test accuracy and faster stable convergence.”
BackgroundSemantic Image Inpainting with Perceptual and Contextual Losses.Preprint
Yeh, R. A., Chen, C., Lim, T.-Y., Hasegawa‐Johnson, M., & Do, M. N. (2016) · arXiv (Cornell University) · Cited by 298 · see citing works
A discriminative deep model for pedestrian detection with occlusion handling
Ouyang, W., & Wang, X. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248062 · Cited by 281 · cited 1× · see citing works
“Recently, deep learning methods have greatly boosted the performance of pedestrian detection [29, 33, 28, 36, 41].”
BackgroundStrengthening the Effectiveness of Pedestrian Detection with Spatially Pooled Features
Paisitkriangkrai, S., Shen, C., & van den Hengel, A. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-10593-2_36 · Cited by 170 · cited 2× · see citing works
“We compare the result of Perceptual GAN with all the existing methods that achieved best performance on the Caltech testing set, including VJ [37], HOG [5], LDCF [27], Katamari [2], SpatialPooling+ [30], TA-CNN [36], Checkerboards [43], CompACT-Deep [44] !" #$%%# &'(”
MethodInfluentialTraffic sign detection based on convolutional neural networks
Wu, Y., Liu, Y., Li, J., Liu, H., & Hu, X. (2013) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2013.6706811 · Cited by 107 · cited 2× · see citing works
“[38] used a CNN combined with fixed and learnable filters to detect traffic signs.”
BackgroundMethodSelf-citationReal Time Traffic Sign Detection Using Color and Shape-Based Features
Le, T., Tran, S. T., Mita, S., et al. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-12101-2_28 · Cited by 46 · cited 2× · see citing works
“Traditional methods for this task includes [20] [15].”
BackgroundMethodA novel pLSA based Traffic Signs Classification SystemPreprint
Haloi, M. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1503.06643 · Cited by 21 · cited 2× · see citing works
“Traditional methods for this task includes [20] [15].”
BackgroundMethod
Flagged first, then by citations
2 of the 48 references couldn’t be matched to a known paper.
Research works printed but not linked: 1
- jiang, 2016, Object detection and counting with low quality videos, Technical Report
Cited by
Sorted by citations
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
Wang, T.-C., Liu, M.-Y., Zhu, J.-Y., Tao, A., Kautz, J., & Catanzaro, B. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00917 · Cited by 4,514 · see citing works
“GANs enable a wide variety of applications such as image generation [1, 41, 60], representation learning [44], image manipulation [62], object detection [32], and video applications [37, 50, 52].”
BackgroundObject Detection in 20 Years: A Survey
Zou, Z., Chen, K., Shi, Z., Guo, Y., & Ye, J. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3238524 · Cited by 3,689 · see citing works
H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes
Li, X., Chen, H., Qi, X., Dou, Q., Fu, C., & Heng, P. (2018) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2018.2845918 · Cited by 2,530 · see citing works
“Recently, perceptual generative adversarial networks (GANs) have been proposed for small object detection and classification [58], [59].”
BackgroundDeep Learning for Image Super-Resolution: A Survey
Wang, Z., Chen, J., & Hoi, S. C. H. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2982166 · Cited by 2,033 · see citing works
Deep Learning in Mobile and Wireless Networking: A Survey
Zhang, C., Patras, P., & Haddadi, H. (2019) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2019.2904897 · Cited by 1,847 · see citing works
“, super-resolution [177], object detection [178], and face completion [179]) by minimizing the divergence between inferred and real data distributions.”
BackgroundFast Underwater Image Enhancement for Improved Visual Perception
Islam, M. J., Xia, Y., & Sattar, J. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.2974710 · Cited by 1,710 · see citing works
“Qualitative performance comparison of FUnIE-GAN and FUnIE-GAN-UP with learning-based methods: CycleGAN [17], UGAN-P [6], Pix2Pix [16], LS-GAN [28], Res-GAN [43], and Res-WGAN [26]; the super-scripted asterisk (∗) denotes unpaired training.”
BackgroundMethodInfluentialA Survey of Deep Learning-Based Object Detection
Jiao, L., Zhang, F., Liu, F., Yang, S., Li, L., Feng, Z., & Qu, R. (2019) · IEEE Access · doi:10.1109/access.2019.2939201 · Cited by 1,392 · see citing works
“[96] utilize GAN model in which generator transfer perceived poor representations of the small objects to super-resolved ones that are similar enough to real large objects to fool a competing discriminator.”
MethodA Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
Gui, J., Sun, Z., Wen, Y., Tao, D., & Ye, J. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3130191 · Cited by 1,231 · see citing works
“[80] proposed perceptual GAN and Bai et al.”
BackgroundImage2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?
Abdal, R., Qin, Y., & Wonka, P. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00453 · Cited by 1,082 · see citing works
“texture synthesis [18, 33, 28], video generation [31, 30], image-to-image translation [11, 36, 1, 24] and object detection [19].”
Recent advances in deep learning for object detection
Wu, X., Sahoo, D., & Hoi, S. C. H. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.01.085 · Cited by 1,047 · see citing works
“[158] proposed a new framework Perceptual GAN for small 1327 object detection.”
BackgroundAugmentation for small object detection
Kisantal, M., Wojna, Z., Murawski, J., Naruniec, J., & Cho, K. (2019) · doi:10.5121/csit.2019.91713 · Cited by 805 · see citing works
“[22] instead uses a Generative Adversarial Network (GAN) to build features in a convolutional network that are indistinguishable between small and large objects in the context of a traffic sign and pedestrian detection.”
MethodTowards Large-Scale Small Object Detection: Survey and Benchmarks
Cheng, G., Yuan, X., Yao, X., Yan, K., Zeng, Q., Xie, X., & Han, J. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3290594 · Cited by 683 · see citing works
“To deal with this issue, Li et al. [121] devised PerceptualGAN to mine and exploit the intrinsic correlations between small-scale and large-scale objects, in which the generator learns to map the weak representations of small objects to super-resolved ones to deceive the discriminator.”
BackgroundDeeply Supervised Salient Object Detection with Short Connections
Hou, Q., Cheng, M., Hu, X., Borji, A., Tu, Z., & Torr, P. H. S. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2815688 · Cited by 674 · see citing works
“In a variety of computer vision tasks, such as image classification [27], [28], semantic segmentation [29], edge detection [26], [30], object detection [31], [32], and pedestrian detection [33], convolutional neural networks (CNNs) [34] have successfully broken the limits of traditional hand-crafted features.”
BackgroundRecent advances in small object detection based on deep learning: A review
Tong, K., Wu, Y., & Zhou, F. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.103910 · Cited by 603 · see citing works
A review of object detection based on deep learning
Xiao, Y., Tian, Z., Yu, J., Zhang, Y., Liu, S., Du, S., & Lan, X. (2020) · Multimedia Tools and Applications · doi:10.1007/s11042-020-08976-6 · Cited by 584 · see citing works
“proposed Perceptual Generative Adversarial Networks (GANs) to improve small object detection [128].”
BackgroundQueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection
Yang, C., Huang, Z., & Wang, N. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01330 · Cited by 634 · see citing works
“These methods can be mainly categorized into four types: 1) increasing the resolution of input features [31, 2, 23, 11, 28, 46, 26, 53]; 2) oversampling and strong data augmentation [31, 21, 75]; 3) incorporating context information [7, 64, 5], and 4) scale-aware training [28, 47, 48, 27].”
MethodDeep learning fault diagnosis method based on global optimization GAN for unbalanced data
Zhou, F., Yang, S., Fujita, H., Chen, D., & Wen, C. (2019) · Knowledge-Based Systems · doi:10.1016/j.knosys.2019.07.008 · Cited by 570 · see citing works
A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds
Wang, Y., Wang, C., Zhang, H., Dong, Y., & Wei, S. (2019) · Remote Sensing · doi:10.3390/rs11070765 · Cited by 529 · see citing works
Zeng, N., Wu, P., Wang, Z., Li, H., Liu, W., & Liu, X. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3153997 · Cited by 538 · see citing works
Image2StyleGAN++: How to Edit the Embedded Images?
Abdal, R., Qin, Y., & Wonka, P. (2020) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr42600.2020.00832 · Cited by 499 · see citing works
“object detection [23], texture synthesis [22, 37, 31], image-to-image translation [16, 42, 28, 25] and video generation [33, 32, 35, 34].”
BackgroundA comprehensive review of object detection with deep learning
Kaur, R., & Singh, S. (2023) · Digital Signal Processing · doi:10.1016/j.dsp.2022.103812 · Cited by 539 · see citing works
Detecting tiny objects in aerial images: A normalized Wasserstein distance and a new benchmark
Xu, C., Wang, J., Yang, W., Yu, H., Yu, L., & Xia, G.-S. (2022) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2022.06.002 · Cited by 504 · see citing works
“While the Generative Adversarial Network (GAN) is proposed to directly generate super-resolved representations for small objects (Li et al., 2017; Bai et al., 2018; Noh et al., 2019).”
Extended Feature Pyramid Network for Small Object Detection
Deng, C., Wang, M., Liu, L., Liu, Y., & Jiang, Y. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3074273 · Cited by 479 · see citing works
“Li et al. [11] use Perceptual GAN to enhance features of small objects with the charac-Authorized teristics of large objects.”
BackgroundMethodInfluentialTiny Object Detection in Aerial Images
Wang, J., Yang, W. T., Guo, H., Zhang, R., & Xia, G.-S. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9413340 · Cited by 470 · see citing works
“Despite the fact that lots of methods tackled this problem [9]–[13], as shown in recent work about the upper bound of object detection performance [14], there is still a large gap between current and the upper bound performances on tiny object detection.”
BackgroundToward Characteristic-Preserving Image-Based Virtual Try-On Network
Wang, B., Zheng, H., Liang, X., Chen, Y., Lin, L., & Yang, M. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01261-8_36 · Cited by 445 · see citing works
“Conditional generative adversarial networks(cGANs) have shown impressive results on image-to-image translation, whose goal is to translate an input image from one domain to another domain [12,38,5,34,18,19,35].”
BackgroundSelf-citationApplications of Generative Adversarial Networks (GANs): An Updated Review
Alqahtani, H., Kavakli, M., & Kumar, G. (2019) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-019-09388-y · Cited by 439 · see citing works
“…high-resolution images from low-resolution ones [11, 53, 60, 115], predicting which drug could treat a certain disease, Object detection [28, 66, 66, 148], retrieving images that contain a given pattern [6, 76, 109, 112], Facial Attribute Manipulation [52, 59, 125, 137, 141], Anime…”
BackgroundInfluentialDeep learning for unmanned aerial vehicle-based object detection and tracking: a survey
Wu, X., Li, W., Hong, D., Tao, R., & Du, Q. (2022) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2021.3115137 · Cited by 421 · see citing works
Perceptual Adversarial Networks for Image-to-Image Transformation
Wang, C., Xu, C., Wang, C., & Tao, D. (2018) · IEEE Transactions on Image Processing · doi:10.1109/tip.2018.2836316 · Cited by 384 · see citing works
“Specifically, [44] devised a novel perceptual discriminator network, which contains an adversarial branch and a perception branch.”
BackgroundImproved YOLOv5 network for real-time multi-scale traffic sign detection
Wang, J., Chen, Y., Dong, Z., & Gao, M. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-08077-5 · Cited by 377 · see citing works
“A novel perceptual generative adversarial network was developed for the recognition of small-sized traffic signs [18], which boosts detection performance by generating super-resolve representations for small traffic signs.”
BackgroundHow Generative Adversarial Networks and Their Variants Work
Hong, Y., Hwang, U., Yoo, J., & Yoon, S. (2019) · ACM Computing Surveys · doi:10.1145/3301282 · Cited by 389 · see citing works
“Notably, Li et al. [69] tries to transform a small object with low resolution into a super resolved large object to make the object more discriminative.”
BackgroundYang, X., Yan, J., Liao, W., Yang, X., Tang, J., & He, T. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3166956 · Cited by 377 · see citing works
“Common small object solutions include data augmentation [49], multi-scale feature fusion [14], [50], tailored sampling strategies [30], [51], [52], generative adversarial networks [53], and multi-scale training [54] etc.”
BackgroundA Survey on Generative Adversarial Networks: Variants, Applications, and Training
Jabbar, A., Li, X., & Omar, B. (2020) · ACM Computing Surveys · doi:10.1145/3463475 · Cited by 387 · see citing works
“nputs and generates the mask of the entire occluded object. Both G and D trained in an adversarial way for the generation of an object image that has invisible regions. For SOD, Perceptual-GAN (PGAN) [78] generates ultra-resolved descriptions of small objects for better detection by narrowing the differences between s…”
BackgroundAtitallah, S. B., Driss, M., Boulila, W., & Ghézala, H. B. (2020) · Computer Science Review · doi:10.1016/j.cosrev.2020.100303 · Cited by 375 · see citing works
Chen, G., Wang, H., Chen, K., Li, Z., Song, Z., Liu, Y., Chen, W., & Knoll, A. (2020) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2020.3005231 · Cited by 364 · see citing works
“The common overflow of GAN-based methods is exhibited in Fig.”
MethodInfluentialDensity Map Guided Object Detection in Aerial Images
Li, C., Yang, T., Zhu, S., Chen, C., & Guan, S. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00103 · Cited by 344 · see citing works
“Early works [9, 2] on aerial image object detection simply leverage the general object detection architecture and focus on improving the detection of small objects.”
BackgroundRFLA: Gaussian Receptive Field Based Label Assignment for Tiny Object Detection
Xu, C., Wang, J., Yang, W. T., Yu, H., Yu, L., & Xia, G.-S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20077-9_31 · Cited by 350 · see citing works
Small Object Detection using Context and Attention
Lim, J.-S., Astrid, M., Yoon, H., & Lee, S. (2021) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic51459.2021.9415217 · Cited by 341 · see citing works
“Recently, several ideas have been proposed for detecting small object [2], [7], [8], [10].”
BackgroundMethodA Normalized Gaussian Wasserstein Distance for Tiny Object DetectionPreprint
Wang, J., Xu, C., Yang, W. T., & Yu, L. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.13389 · Cited by 500 · see citing works
“GAN-based Detectors: Perceptual GAN [12] is the first to attempt to apply GAN to small object detection, it improves small object detection through narrowing representation difference of small objects from the large ones.”
InfluentialEuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes
Braun, M., Krebs, S., Flohr, F., & Gavrila, D. M. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2897684 · Cited by 307 · see citing works
“For Faster R-CNN, R-FCN and SSD we use VGG-16 [45] as base network as it is very common in pedestrian detection [4], [60], [62], [63], [71].”
MethodSynthetic data augmentation for surface defect detection and classification using deep learning
Jain, S., Seth, G., Paruthi, A., Soni, U., & Kumar, G. (2022) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-020-01710-x · Cited by 302 · see citing works
“…applications of GANs include handwriting recognition (Tian 2017), facial age progression (Zhang et al. 2017), super-resolution imaging (Ledig et al. 2017), visual saliency prediction (Pan et al. 2017), object detection (Li et al. 2017), and unsupervised domain adaptation (Bous-malis et al. 2017).”
BackgroundRevisiting RCNN: On Awakening the Classification Power of Faster RCNN
Cheng, B., Wei, Y., Shi, H., Feris, R., Xiong, J., & Huang, T. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01267-0_28 · Cited by 282 · see citing works
“We also compare the influence of size of sampled RoIs during training: [8, 16, 32, 64].”
Self-citationNoh, J., Bae, W., Lee, W., Seo, J., & Kim, G. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00982 · Cited by 271 · see citing works
Joint Pose and Expression Modeling for Facial Expression Recognition
Zhang, F., Zhang, T., Mao, Q., & Xu, C. (2018) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2018.00354 · Cited by 266 · see citing works
“Besides, during the last three years, several approaches [4, 65, 55, 25, 29] have been proposed to improve the original GAN from different perspectives.”
BackgroundVisual Perception Enabled Industry Intelligence: State of the Art, Challenges and Prospects
Yang, J., Wang, C., Jiang, B., Song, H., & Meng, Q. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.2998818 · Cited by 268 · see citing works
“portant application fields, such as image generation [95], video generation and prediction [96], object detection [97], image translation [98], image editing [99], image restoration [100],”
BackgroundA survey on generative adversarial networks for imbalance problems in computer vision tasks
Sampath, V., Maurtua, I., Martín, J. J. A., & Gutierrez, A. (2021) · Journal Of Big Data · doi:10.1186/s40537-021-00414-0 · Cited by 275 · see citing works
“Commonly adopted strategy is to convert low resolution small object features into high resolution features [208] using GAN.”
BackgroundMethodInfluentialAn Improved Faster R-CNN for Small Object Detection
Cao, C. C., Wang, B., Zhang, W., Zeng, X., Yan, X., Feng, Z., Liu, Y., & Wu, Z. (2019) · IEEE Access · doi:10.1109/access.2019.2932731 · Cited by 262 · see citing works
A Review on Generative Adversarial Networks: Algorithms, Theory, and ApplicationsPreprint
Gui, J., Sun, Z., Wen, Y., Tao, D., & Ye, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.06937 · Cited by 252 · see citing works
“[80] proposed perceptual GAN and Bai et al.”
BackgroundLightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
Zhang, M., Xu, S., Song, W., He, Q., & Wei, Q. (2021) · Remote Sensing · doi:10.3390/rs13224706 · Cited by 237 · see citing works
“At present, multi-scale feature learning, data augmentation [50,51] or feature fusion are the most popular research methods for small object detection.”
Bosquet, B., Cores, D., Seidenari, L., Brea, V. M., Mucientes, M., & Bimbo, A. D. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108998 · Cited by 232 · see citing works
“A way to increase small object detection accuracy is to improve object resolution, for example with a Perceptual GAN [21] or an SOD-MTGAN [22] .”
MethodSemantics Disentangling for Text-To-Image Generation
Yin, G., Liu, B., Sheng, L., Yu, N., Wang, X., & Shao, J. (2019) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2019.00243 · Cited by 218 · see citing works
“GAN-based algorithms show excellent performance in image generation [21, 1, 20, 25, 35, 2, 23].”
BackgroundSRFeat: Single Image Super-Resolution with Feature Discrimination
Park, S., Son, H., Cho, S., Hong, K.-S., & Lee, S. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01270-0_27 · Cited by 209 · see citing works
“Thanks to its effectiveness in image generation, GAN has been widely applied to various tasks such as image synthesis, style transfer, image inpainting, and object detection [37, 55, 20, 25, 23, 19, 30, 28].”
MethodHRDNet: High-Resolution Detection Network for Small Objects
Liu, Z., Gao, G., Sun, L., & Fang, Z. (2021) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme51207.2021.9428241 · Cited by 209 · see citing works
“Perceptual GAN [13] generated super-resolved features and stacked them into feature maps of small objects to enhance the representations.”
BackgroundYOLOv8-QSD: An Improved Small Object Detection Algorithm for Autonomous Vehicles Based on YOLOv8
Wang, H., Liu, C., Cai, Y., Chen, L., & Li, Y. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3379090 · Cited by 205 · see citing works
Tools, techniques, datasets and application areas for object detection in an image: a review
Kaur, J., & Singh, W. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13153-y · Cited by 194 · see citing works
“Some new approaches, such as adversarial training, different backbone architecture, and attention mechanism, have been used to increase speed or improve detection processes in challenging and complex environments [66, 72, 88, 100, 109, 135].”
BackgroundVSSA-NET: Vertical Spatial Sequence Attention Network for Traffic Sign Detection
Yuan, Y., Xiong, Z., & Wang, Q. (2019) · IEEE Transactions on Image Processing · doi:10.1109/tip.2019.2896952 · Cited by 190 · see citing works
Recent Advances of Generative Adversarial Networks in Computer Vision
Cao, Y., Jia, L., Chen, Y.-X., Lin, N., Yang, C., Zhang, B., Liu, Z., Li, X., & Dai, H. (2018) · IEEE Access · doi:10.1109/access.2018.2886814 · Cited by 191 · see citing works
“For example, perceptual generative adversarial network [92] is for small object detection by reducing the representation difference between small and large objects.”
BackgroundVehicle Detection From UAV Imagery With Deep Learning: A Review
Bouguettaya, A., Zarzour, H., Kechida, A., & Taberkit, A. M. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3080276 · Cited by 187 · see citing works
“GAN architecture can enhance the poor representations of tiny objects to super-resolved ones in the generator part to fool the discriminator part, which could improve small vehicle detection performance from high-altitude UAV imagery [45].”
BackgroundVision-Based Traffic Sign Detection and Recognition Systems: Current Trends and Challenges
Wali, S. B., Abdullah, M. A., Hannan, M. A., Hussain, A., Samad, S. A., Ker, P. J., & Mansor, M. (2019) · Sensors · doi:10.3390/s19092093 · Cited by 183 · see citing works
“Of the examples applied to traffic sign classification are committee CNN [113], multi-scale CNN [114], multi-column CNN [102], multi-task CNN [111,115], hinge-loss CNN [116], deep CNN [46,117], a CNN with diluted convolutions [118], a CNN with a generative adversarial network (GAN) [119], and a CNN with SVM [120].”
MethodDynamic Coarse-to-Fine Learning for Oriented Tiny Object Detection
Xu, C., Ding, J., Wang, J., Yang, W. T., Yu, H., Yu, L., & Xia, G.-S. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00707 · Cited by 183 · see citing works
“PGAN [25] first applies GAN to small object detection.”
BackgroundA comprehensive review of deep learning-based single image super-resolution
Bashir, S. M. A., Wang, Y., Khan, M., & Niu, Y. (2021) · PeerJ Computer Science · doi:10.7717/peerj-cs.621 · Cited by 174 · see citing works
“Other applications of SR include object detection [226,227], stereo image SR [228–230]; overall, SR plays a vital role in multi-disciplines from medical science, computer vision to satellite imaging and remote sensing.”
BackgroundRetractedAn improved Yolov5 real-time detection method for small objects captured by UAV
Zhan, W., Sun, C., Wang, M., She, J., Zhang, Y., Zhang, Z., & Sun, Y. (2021) · Soft Computing · doi:10.1007/s00500-021-06407-8 · Cited by 171 · see citing works
“The third is generative adversarial networks (GANs) (Li et al. 2017a).”
BackgroundLearning to Generate Time-Lapse Videos Using Multi-stage Dynamic Generative Adversarial Networks
Xiong, W., Luo, W., Ma, L., Liu, W., & Luo, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00251 · Cited by 162 · see citing works
“Although the perceptual loss performs well in combination with GANs [13, 14] on some tasks, it typically requires features to be extracted from a pretrained deep neural network, which is both time and space consuming.”
BackgroundFrom Handcrafted to Deep Features for Pedestrian Detection: A Survey
Cao, J., Pang, Y., Xie, J., Khan, F. S., & Shao, L. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3076733 · Cited by 164 · see citing works
“Scale-aware Part-aware Context Description DDN [138] CVPR2012 Hybrid SW HOG softmax NMS no yes no first deep model for pedestrian detection UMS [169] CVPR2013 P-CNN SW CNN softmax NMS no no ms fusion one of the earliest deep pedestrian detectors UDN [139] ICCV2013 P-CNN SW CNN softmax NMS no yes no join different comp…”
BackgroundFloW: A Dataset and Benchmark for Floating Waste Detection in Inland Waters
Cheng, Y., Zhu, J., Jiang, M., Fu, J., Pang, C., Wang, P., Sankaran, K., Onabola, O., Liu, Y., Liu, D., et al. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01077 · Cited by 162 · see citing works
“For object detection based on deep neural network, compared to large objects, small objects usually lack appearance information and the high-level features of small objects are not discriminative, making it hard to distinguish them from the background and achieve accurate localization [2, 22].”
BackgroundSmall Object Detection via Coarse-to-fine Proposal Generation and Imitation Learning
Yuan, X., Cheng, G., Yan, K., Zeng, Q., & Han, J. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00581 · Cited by 159 · see citing works
“Based on the generative adversarial paradigm, Perceptual GAN [24] designs a generator that is optimized to produce high-quality representations of small instances to fool the subsequent discriminator.”
BackgroundMethodInfluentialJi, S.-J., Ling, Q.-H., & Han, F. (2022) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2022.108490 · Cited by 150 · see citing works
Selective synthetic augmentation with HistoGAN for improved histopathology image classification
Xue, Y., Ye, J., Zhou, Q., Long, L. R., Antani, S., Xue, Z., Cornwell, C., Zaino, R. J., Cheng, K. C., & Huang, X. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101816 · Cited by 146 · see citing works
“Generative adversarial networks (GANs) [11] as an unsupervised learning technique, has enabled a wide variety of applications including image synthesis, object detection [23] and image segmentation [44].”
BackgroundQi, G., Zhang, Y., Wang, K., Mazur, N., Liu, Y., & Malaviya, D. (2022) · Remote Sensing · doi:10.3390/rs14020420 · Cited by 142 · see citing works
“Partial experimental data shown in this table are missing, because some models [15,33,45,50]”
BackgroundMethodInfluentialXing, H., Gao, S., Wang, Y., et al. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3255304 · Cited by 141 · see citing works
A Conditional Adversarial Network for Semantic Segmentation of Brain Tumor
Rezaei, M., Harmuth, K., Gierke, W., Kellermeier, T., Fischer, M., Yang, H., & Meinel, C. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-75238-9_21 · Cited by 138 · see citing works
YOLC: You Only Look Clusters for Tiny Object Detection in Aerial Images
Liu, C., Gao, G., Huang, Z., Hu, Z., Liu, Q., & Wang, Y. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3386928 · Cited by 146 · see citing works
“Cheng et al. [30] provide an extensive review of small object detection methods, categorizing them into various groups such as sample-oriented [10], [31], scale-aware [6], [7], [32], [33], super-resolution-based [8], [34], context modeling [35], and focus-and-detect [12], [13], [36].”
BackgroundCoarse-grained Density Map Guided Object Detection in Aerial Images
Duan, C., Wei, Z., Zhang, C., Qu, S., & Wang, H. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00313 · Cited by 136 · see citing works
“Perceptual GAN [14] leveraged a confrontation generation network to generate super-resolution images of small objects, thereby improving the detection ability of small objects.”
BackgroundSCH-GAN: Semi-Supervised Cross-Modal Hashing by Generative Adversarial Network
Zhang, J., Peng, Y., & Yuan, M. (2018) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2018.2868826 · Cited by 133 · see citing works
“Inspired by the GAN [31]–[34], the adversarial process is defined as follows:”
BackgroundMethodInfluentialReal-time gun detection in CCTV: An open problem
González, J. L. S., Zaccaro, C., Álvarez-García, J. A., Morillo, L. M. S., & Caparrini, F. S. (2020) · Neural Networks · doi:10.1016/j.neunet.2020.09.013 · Cited by 130 · see citing works
“This problem has been previously approached by modifying the network architecture to apply object detection in aerial and satellite images [3, 4], increasing image resolution by adding Perceptual Generative Adversarial Networks [5], or manually adding small objects to the scenes [6].”
MethodCFANet: Efficient Detection of UAV Image Based on Cross-Layer Feature Aggregation
Zhang, Y., Wu, C., Guo, W., Zhang, T., & Li, W. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3273314 · Cited by 132 · see citing works
Research on License Plate Recognition Algorithms Based on Deep Learning in Complex Environment
Weihong, W., & Jiaoyang, T. (2020) · IEEE Access · doi:10.1109/access.2020.2994287 · Cited by 130 · see citing works
“Li et al. [32] was used Perceptual GAN to internally lift representations of small objects to ‘‘super-resolved’’ ones, achieving similar characteristics as large objects and thus more discriminative for detection.”
Automatic Crack Segmentation for UAV-Assisted Bridge Inspection
Ayele, Y. Z., Aliyari, M., Griffiths, D., & Droguett, E. L. (2020) · Energies · doi:10.3390/en13236250 · Cited by 128 · see citing works
Adversarial Discriminative Heterogeneous Face Recognition
Song, L., Zhang, M., Wu, X., et al. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12291 · Cited by 124 · see citing works
“2017) , saliency detection (Hu, Zhao, and Tan 2017) and object detection (Li et al. 2017; Wang, Shrivastava, and Gupta 2017).”
BackgroundA review of small object detection based on deep learning
Wei, W., Cheng, Y., He, J., & Zhu, X. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-09422-6 · Cited by 127 · see citing works
“Li et al. [45] proposed a perceptual GAN method specifically for small object detection in the GAN-based approach.”
Dynamic High-Pass Filtering and Multi-Spectral Attention for Image Super-Resolution
Magid, S. A., Zhang, Y., Wei, D., Jang, W.-D., Lin, Z., Fu, Y., & Pfister, H. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00425 · Cited by 124 · see citing works
“Given the recent advances in deep learning, convolutinal neural network (CNN) based SR methods have been leveraged in a wide variety of research domains such as biomedicine, object recognition, and hyper-spectral imaging [9, 21, 32, 43].”
BackgroundSmall Object Detection Based on Deep Learning for Remote Sensing: A Comprehensive Review
Wang, X., Wang, A., Yi, J., Song, Y., & Chehri, A. (2023) · Remote Sensing · doi:10.3390/rs15133265 · Cited by 124 · see citing works
“GAN-based PGAN [104] and SOD-MTGAN [105] inherit the generator and discriminator.”
A geometric view of optimal transportation and generative model
Lei, N., Su, K., Cui, L., Yau, S., & Gu, X. D. (2018) · Computer Aided Geometric Design · doi:10.1016/j.cagd.2018.10.005 · Cited by 121 · see citing works
“…to numerous computer vision tasks such as image inplainting [35, 49, 28], image super resolution [26, 19], semantic segmentation [52, 31], object detection [37, 27, 47], video prediction [32, 46], image translation [20, 51, 7, 29], 3D vision [48, 34], face editing [25, 30, 36, 39, 6, 40, 18], etc.”
BackgroundImaGINator: Conditional Spatio-Temporal GAN for Video Generation
Wang, Y., Biliński, P., Brémond, F., & Dantcheva, A. (2020) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv45572.2020.9093492 · Cited by 122 · see citing works
“Conditional generation has been beneficial in domain transfer, super-resolution imaging, video to video translation, as well as image and face editing [13, 54, 26, 15, 20, 43, 4, 16, 45, 46, 50, 44].”
BackgroundContext-Aware Block Net for Small Object Detection
Cui, L., Lv, P., Jiang, X., et al. (2020) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2020.3004636 · Cited by 120 · see citing works
Wang, F., Wang, R., Xie, C., Yang, P., & Liu, L. (2020) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2020.105222 · Cited by 120 · see citing works
“Li et al. (2017) attempt to solve the small object detection problem by present a Perceptual Generative Adversarial Network (PGAN), which generate high resolution image by low resolution ones and obtain similar features and hence improve the detection rate.”
BackgroundReal-Time Detection Method for Small Traffic Signs Based on Yolov3
Zhang, H., Qin, L., Li, J., et al. (2020) · IEEE Access · doi:10.1109/access.2020.2984554 · Cited by 115 · see citing works
“Li et al. [29] proposed a Perceptual Generative Adversarial Network (Perceptual GAN) model that improves the detection accuracy of small traffic signs in the TT100K data set through narrowing the representation difference of small objects from large objects.”
BackgroundSelf-citationImproved generative adversarial network for vibration-based fault diagnosis with imbalanced data
Zhao, B., & Yuan, Q. (2020) · Measurement · doi:10.1016/j.measurement.2020.108522 · Cited by 113 · see citing works
Karthi, M., Muthulakshmi, V., Priscilla, R., et al. (2021) · International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES) · doi:10.1109/icses52305.2021.9633834 · Cited by 113 · see citing works
RetractedHan, W., Chen, J., Wang, L., Feng, R., Li, F., Wu, L., Tian, T., & Yan, J. (2021) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2020.3041450 · Cited by 115 · see citing works
An Adversarial Approach to Hard Triplet Generation
Zhao, Y., Jin, Z., Qi, G.-J., Lu, H., & Hua, X. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01240-3_31 · Cited by 109 · see citing works
“[15] proposed Perceptual GANs that generate super-resolution representations for small objects to boost detection accuracy by leveraging iteratively updated generator networks and discriminator networks.”
BackgroundRecent Advances in Object Detection in the Age of Deep Convolutional Neural NetworksPreprint
Agarwal, S., du Terrail, J. O., & Jurie, F. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1809.03193 · Cited by 133 · see citing works
“Light-head R-CNN [Li et al., 2017f] proposed a smaller detection head and thin feature maps to speed up two-stage detectors.”
BackgroundInfluentialMulti-pathway generative adversarial hashing for unsupervised cross-modal retrieval
Zhang, J., & Peng, Y. (2020) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2019.2922128 · Cited by 103 · see citing works
“Besides image synthesis, GAN is also applied to video prediction [52] and object detection [53].”
MethodGuan, S., & Loew, M. (2019) · Journal of medical imaging · doi:10.1117/1.jmi.6.3.031411 · Cited by 109 · see citing works
Attentional feature pyramid network for small object detection
Min, K., Lee, G., & Lee, S. (2022) · Neural Networks · doi:10.1016/j.neunet.2022.08.029 · Cited by 105 · see citing works
Cascaded Zoom-in Detector for High Resolution Aerial Images
Meethal, A., Granger, É., & Pedersoli, M. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00198 · Cited by 107 · see citing works
“[5] provides a comprehensive survey of small object detection techniques which is broadly classified into: scale aware training [3,11, 20, 21, 25, 31, 32, 41], super-resolution methods [19], context modeling [1], and density guided detection [9,18,39] among others.”
BackgroundPang, Y., Cao, J., Wang, J., & Han, J. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2916592 · Cited by 99 · see citing works
“More importantly, the results show that our JCS-Net is superior to PGAN.”
MethodInfluentialWu, Z., Suresh, K., Narayanan, P., Xu, H., Kwon, H., & Wang, Z. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00129 · Cited by 99 · see citing works
“[47, 27, 30] proposed to internally super-resolve the feature maps of small objects to make them resemble similar characteristics as large objects.”
BackgroundThe Mapillary Traffic Sign Dataset for Detection and Classification on a Global Scale
Ertler, C., Mislej, J., Ollmann, T., Porzi, L., Neuhold, G., & Kuang, Y. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58592-1_5 · Cited by 106 · see citing works
“Traffic sign detection and classification have been studied extensively in computer vision [14,20,25,35].”
BackgroundGąsienica-Józkowy, J., Knapik, M., & Cyganek, B. (2021) · Integrated Computer-Aided Engineering · doi:10.3233/ica-210649 · Cited by 100 · see citing works
“Different approaches were suggested by Li [50] and by Bai [51], both proposing generative adversarial networks (GAN) to increase feature-map resolution, in result increasing performance of further detection modules.”
Spectral–Spatial Self-Attention Networks for Hyperspectral Image Classification
Zhang, X., Sun, G., Jia, X., Wu, L., Zhang, A., Ren, J., Fu, H., & Yao, Y. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3102143 · Cited by 98 · see citing works
- Authors
- Li, J., Liang, X., Wei, Y., Xu, T., Feng, J., & Yan, S.
- Affiliations
- Beijing Institute of Technology (CN)
- Carnegie Mellon University (US)
- National University of Singapore (SG)
- Venue
- 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2017-07
- Pages
- 1951-1959
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2625219738
- Semantic Scholar
- 75e02f14e9871a405d2d367dd7c77c730499fed6
- Cited by
- 953
- References
- 48
- Citations by year
- 2026: 40 · 2025: 104 · 2024: 136 · 2023: 135 · 2022: 124 · 2021: 143 · 2020: 103 · 2019: 99 · 2018: 61 · 2017: 8
- FWCI
- 21.67
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Tencent
- National University of Singapore (R-263-000-C08-133)
- Ministry of Education - Singapore
- China Scholarship Council (201506030045)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Generative Adversarial Networks and Image Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- small object detection, super-resolution, representation transfer, perceptual loss, traffic sign detection, pedestrian detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
