Convolutional networks and applications in vision
Open accessThe DOI matches its Crossref record; 52 of the 56 references it cites were found in a registry or an index.
Abstract
Intelligent tasks, such as visual perception, auditory perception, and language understanding require the construction of good internal representations of the world (or "features")? which must be invariant to irrelevant variations of the input while, preserving relevant information. A major question for Machine Learning is how to learn such good features automatically. Convolutional Networks (ConvNets) are a biologically-inspired trainable architecture that can learn invariant features. Each stage in a ConvNets is composed of a filter bank, some nonlinearities, and feature pooling layers. With multiple stages, a ConvNet can learn multi-level hierarchies of features. While ConvNets have been successfully deployed in many commercial applications from OCR to video surveillance, they require large amounts of labeled training samples. We describe new unsupervised learning algorithms, and new non-linear stages that allow ConvNets to be trained with very few labeled samples. Applications to visual object recognition and vision navigation for off-road mobile robots are described.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The text we found for this paper looks like another document, so the self-check wasn’t run.
Citations
Cites
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
Self-citationDistinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589 · cited 3×
“Most common feature extraction systems like SIFT [40], HoG [41] are composed of filterbanks (oriented edge detectors at multiple scales) followed by non-linearities (winner take all) and pooling (histogramming).”
MethodHistograms 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×
“Most common feature extraction systems like SIFT [40], HoG [41] are composed of filterbanks (oriented edge detectors at multiple scales) followed by non-linearities (winner take all) and pooling (histogramming).”
MethodReducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Backpropagation Applied to Handwritten Zip Code Recognition
LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.4.541 · Cited by 11,891 · cited 1×
“The innovation in [5], [1] was to simplify the architecture and to use the back-propagation algorithm to train the entire system in a supervised fashion.”
MethodSelf-citationBeyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
Lazebnik, S., Schmid, C., & Ponce, J. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.68 · Cited by 7,916 · cited 4×
“A Pyramid Match Kernel (PMK) SVM [39] classifer can also be seen as another layer of feature extraction since it performs a K-means based feature extraction followed by local histogramming.”
MethodInfluentialGreedy layer-wise training of deep networks
Bengio, Y., Lamblin, P., Popovici, D., & Larochelle, H. (2007) · Advances in Neural Information Processing Systems 19 · doi:10.7551/mitpress/7503.003.0024 · Cited by 4,841 · cited 1×
“Recent research in deep learning [30], [31], [32] has shown that unsupervised learning can be used to train each stage one after the other using only unlabeled data, reducing the requirement for labeled samples significantly.”
Background(1989) · neural information processing systems · Cited by 3,736
Self-citationSparse coding with an overcomplete basis set: A strategy employed by V1?
Olshausen, B. A., & Field, D. J. (1997) · Vision Research · doi:10.1016/s0042-6989(97)00169-7 · Cited by 3,642
Best practices for convolutional neural networks applied to visual document analysis
Simard, P. Y., Steinkraus, D., & Platt, J. C. (2005) · Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings. · doi:10.1109/icdar.2003.1227801 · Cited by 2,741
Research works printed but not linked: 4
- lecun, 0, Handwritten digit recognition with a backpropagation network, NIPS'89
- frome, 0, Large-scale privacy protection in street-level imagery, ICCV'09
- collobert, 2008, Torch, Workshop on Machine Learning Open Source Software NIPS
- lecun, 2002, The Lush Reference Manual
Cited by
Items 601–700 of 1,001
Multi‐stream 3D CNN structure for human action recognition trained by limited data
Chenarlogh, V. A., & Razzazi, F. (2018) · IET Computer Vision · doi:10.1049/iet-cvi.2018.5088 · Cited by 40
“Owing to fewer shared connections in the first two types of layers, CNNs are faster from traditional neural networks in the training stage [9].”
BackgroundDuggento, A., Aiello, M., Cavaliere, C., Cascella, G. L., Cascella, D., Conte, G., Guerrisi, M., & Toschi, N. (2019) · Contrast Media & Molecular Imaging · doi:10.1155/2019/5982834 · Cited by 40
“In this context, the disruptive innovation in computer vision brought about through what is known as deep learning [15–17], and in particular, a class of methods known as deep convolutional neural networks (CNNs) [18] is very quickly making its way into the world of medical imaging.”
BackgroundQiao, W., & Bi, X. (2019) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2019.101745 · Cited by 40
Deep Learning Acceleration with Neuron-to-Memory Transformation
Imani, M., Samragh, M., Kim, Y., Gupta, S., Koushanfar, F., & Rosing, T. (2020) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca47549.2020.00011 · Cited by 39
“In particular, deep neural networks (DNNs) demonstrate superior effectiveness for diverse classification problems, image processing, video segmentation, speech recognition, computer vision, and gaming [2,3,4,5].”
BackgroundChen, T., Hu, N., Niu, R., Na, Z., & Plaza, A. (2020) · Remote Sensing · doi:10.3390/rs12233895 · Cited by 40
“, visual recognition [23], image retrieval [24], and scene annotation [25].”
BackgroundOptimized deep learning model for mango grading: Hybridizing lion plus firefly algorithm
Tripathi, M. K., & Maktedar, D. D. (2021) · IET Image Processing · doi:10.1049/ipr2.12163 · Cited by 40
“In an image, the spatial information usage among the pixels is the core intention of CNN [31] and therefore it depends on discrete convolution.”
BackgroundMethodBalasubramanian, K., Ananthamoorthy, N. P., & Ramya, K. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07279-1 · Cited by 40
“The CNN automatically extracts the implicit features of patterns representing the inputs and further classifies the patterns from the features extracted [45, 46].”
BackgroundEnabling efficient deep convolutional neural network-based sensor fusion for autonomous driving
Zeng, X., Wang, Z., & Hu, Y. (2022) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3489517.3530444 · Cited by 40
“Typically, DCNN is a stacked structure composed of multiple layers [5].”
Jariwala, N., Putta, C. L., Gatade, K., Umarji, M., Rahman, S. N. R., Pawde, D. M., Sree, A., Kamble, A. S., Goswami, A., Chakraborty, P., et al. (2023) · Journal of Drug Delivery Science and Technology · doi:10.1016/j.jddst.2023.104751 · Cited by 40
On the Impact of Imbalanced Data in Convolutional Neural Networks Performance
Pulgar, F. J., Rivera, A. J., Charte, F., & del Jesús, M. J. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-59650-1_19 · Cited by 40
Doppler-Radar Based Hand Gesture Recognition System Using Convolutional Neural Networks
Zhang, J., Tao, J., & Shi, Z. (2018) · Lecture notes in electrical engineering · doi:10.1007/978-981-10-6571-2_132 · Cited by 40
Gao, C., Zhang, Q., Tan, Z., Zhao, G., Gao, S., Kim, E., & Shen, T. (2024) · Heritage Science · doi:10.1186/s40494-024-01144-1 · Cited by 39
Video action recognition collaborative learning with dynamics via PSO-ConvNet Transformer
Phong, N. H., & Ribeiro, B. (2023) · Scientific Reports · doi:10.1038/s41598-023-39744-9 · Cited by 39
CryptoKnight: Generating and Modelling Compiled Cryptographic Primitives
Hill, G., & Bellekens, X. (2018) · Information · doi:10.3390/info9090231 · Cited by 39
“A variety of substantiated CNN architectures have been used to great effect in computer vision [18] and even Natural Language Processing (NLP), with empirically distinguished superiority in semantic matching [19], compared to other models.”
ResultMontalbo, F. J. P., & Hernandez, A. A. (2020) · IEEE International Colloquium on Signal Processing & Its Applications (CSPA) · doi:10.1109/cspa48992.2020.9068683 · Cited by 39
“The supervised approach of having a multi-stage training process made significant contributions to computer vision aspects in recent years [9].”
BackgroundRao, J., Yang, F., Mo, H., Kollmannsberger, S., & Rank, E. (2022) · Journal of Sound and Vibration · doi:10.1016/j.jsv.2022.117418 · Cited by 39
“In deep learning, a convolutional neural network (CNN), which is one of the most popular choices of deep neural networks [25], is able to approximate the non-linear mapping from the input space to the output space.”
BackgroundA New View of Multisensor Data Fusion: Research on Generalized Fusion
Guo, C., Liu, Z., Yu, G., & Liang, J. (2021) · Mathematical Problems in Engineering · doi:10.1155/2021/5471242 · Cited by 39
Invariant visual object recognition: biologically plausible approaches
Robinson, L., & Rolls, E. T. (2015) · Biological Cybernetics · doi:10.1007/s00422-015-0658-2 · Cited by 80
“Both approaches select filter analysers using feedforward processing without a teacher, in contrast to convolutional and deep learning networks (LeCun et al. 2010) which typically use errors from a teacher backpropagated through multiple layers that do not aim for biological plausibility (Rolls 2008, 2016).”
BackgroundMethodInfluentialMulti-Class Multi-Scale Series Contextual Model for Image Segmentation
Seyedhosseini, M., & Taşdizen, T. (2013) · IEEE Transactions on Image Processing · doi:10.1109/tip.2013.2274388 · Cited by 39
“like maximum pooling can be used instead of Gaussian averaging [34].”
MethodHartmann, K. G., Schirrmeister, R. T., & Ball, T. (2018) · International Winter Conference on Brain-Computer Interface (BCI) · doi:10.1109/iww-bci.2018.8311493 · Cited by 39
“The following 4 convolutional layers are each followed by a layer with exponential linear units [17] and a max-pooling layer [18].”
BackgroundHasan, M. K., Alam, M. A., Elahi, M. T. E., Roy, S., & Wahid, S. R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.11993 · Cited by 39
“Three comparatively shallow networks, such as MobileNetV2, SqueezeNet (LeCun et al., 2010), and ResNet-18, and five deep networks, such as InceptionV3, ResNet-101, CheXNet (Rajpurkar et al., 2018), VGG-19, and DenseNet-201, were CapsNets are alternative models of CNN, which are capable of capturing spatial information…”
BackgroundOptimal Ensemble learning model for COVID-19 detection using chest X-ray images
Balasubramaniam, S., & Kumar, K. S. (2022) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2022.104392 · Cited by 38
“Proposed Ensemble + SAKHO FMHCFDLF [33] DBN [48] FCM [42] LSTM [45] Ensemble + KHO [47] Ensemble + EHO [49] Ensemble + WOA [50] Ensemble + BOA [51] Best 0.”
BackgroundMethodGuida, C., Zhang, M., & Shan, J. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08214-8 · Cited by 38
“A CNN is a machine learning method that can be trained to automate common computer vision tasks such as image classification [18] and segmentation [35].”
MethodIntroduction to deep learning methods for multi‐species predictions
Hu, Y., Si‐Moussi, S., & Thuiller, W. (2024) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14466 · Cited by 38
“For example, convolutional neural networks (CNNs) (He et al., 2016 ; LeCun et al., 2010 ) are particularly well-suited for image data due to their consideration of translation invariance and equivariance, along with parameter efficiency when extracting features from large images.”
A Preliminary Study on Breast Cancer Risk Analysis Using Deep Neural Network
Sun, W., Tseng, T.-L., Zheng, B., & Qian, W. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-41546-8_48 · Cited by 38
MSST-ResNet: Deep multi-scale spatiotemporal features for robust visual object tracking
Liu, B., Liu, Q., Zhu, Z., Zhang, T., & Yang, Y. (2018) · Knowledge-Based Systems · doi:10.1016/j.knosys.2018.10.044 · Cited by 38
In Situ Monitoring of Thin-Wall Build Quality in Laser Powder Bed Fusion Using Deep Learning
Gaikwad, A., Imani, F., Yang, H. Y., Reutzel, E. W., & Rao, P. (2019) · Smart and Sustainable Manufacturing Systems · doi:10.1520/ssms20190027 · Cited by 38
LPI Radar Waveform Recognition Based on Multi-Resolution Deep Feature Fusion
Ni, X., Wang, H., Meng, F., Hu, J. F., & Tong, C. (2021) · IEEE Access · doi:10.1109/access.2021.3058305 · Cited by 38
“Recently, with the successful application of deep learning in image classification and other fields [4], [5], scholars are devoted to converting one-dimensional radar signal into two-dimensional feature images by signal processing and The associate editor coordinating the review of this manuscript…”
BackgroundRolls, E. T. (2021) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2021.686239 · Cited by 38
“…rule in which the information is present in the pre-and postsynaptic rates without the need for backpropagation of error as in deep learning (LeCun et al., 2010, 2015; Yamins and DiCarlo, 2016) or for lateral propagation of synaptic weights as in deep convolution networks (Yamins and…”
BackgroundInfluentialSAR Unlabeled Target Recognition Based on Updating CNN With Assistant Decision[-.2pc]
Cui, Z., Tang, C., Cao, Z., & Dang, S. (2018) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2018.2851600 · Cited by 38
Fainekos, G., Hoxha, B., & Sankaranarayanan, S. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32079-9_3 · Cited by 38
“They rely heavily on machine learning components such as neural networks for detecting other vehicles, obstacles, pedestrians, traffic signals and signs from a combination of image and LiDAR data [38].”
BackgroundA Labeling-Free Approach to Supervising Deep Neural Networks for Retinal Blood Vessel SegmentationPreprint
Chen, Y. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1704.07502 · Cited by 38
Understanding Early Word Learning in Situated Artificial AgentsPreprint
Hill, F., Clark, S. R. L., Hermann, K. M., & Blunsom, P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1710.09867 · Cited by 38
“It is well established that convolutional networks trained to classify images also exhibit differential specialisation of feature detectors between layers (see e.g. LeCun et al. (2010)).”
BackgroundThe Goldilocks Zone: Towards Better Understanding of Neural Network Loss Landscapes
Fort, S., & Scherlis, A. (2019) · AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) · doi:10.1609/aaai.v33i01.33013574 · Cited by 37
Markov Random Field Based Convolutional Neural Networks for Image Classification
Peng, Y., & Yin, H. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-68935-7_42 · Cited by 37
Pereira, S. R. B., Buldain-Pérez, J. D., Comech, M. P., & Gilbert, I. P. (2020) · IEEE Transactions on Power Delivery · doi:10.1109/tpwrd.2020.3042934 · Cited by 37
A Two-Branch CNN Fusing Temporal and Frequency Features for Motor Imagery EEG Decoding
Yang, J. L., Gao, S., & Shen, T. (2022) · Entropy · doi:10.3390/e24030376 · Cited by 38
“, the feature extraction layer formed by the combination of convolution layers and pooling layers, and the classification output layer composed of several fully connected layers [33].”
BackgroundConvolutional Visual Prompt for Robust Visual Perception
Tsai, Y.-Y., Mao, C., & Yang, J. (2023) · neural information processing systems · doi:10.52202/075280-1211 · Cited by 37
“We then apply the convolutional visual prompts on the corrupted samples, which instruct the model to adapt and produce a small self-supervision loss. that convolutional operations are effective for handling structured data with local motifs [30, 64].”
MethodInfluentialFrom music audio to chord tablature: Teaching deep convolutional networks toplay guitar
Humphrey, E. J., & Bello, J. P. (2014) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2014.6854952 · Cited by 37
“There are two primary benefits to using convolutional layers in a deep network; the number of parameters to learn is greatly reduced because only a local neighborhood, rather than the entire input, is processed by each kernel; and two, the kernel is translated across all shifts of the input, allowing it to encode the…”
BackgroundMethodTorres, A. D., Yan, H., Aboutalebi, A. H., Das, A., Duan, L., & Rad, P. (2018) · Elsevier eBooks · doi:10.1016/b978-0-12-813314-9.00003-7 · Cited by 37
Defect Detection in Porcelain Industry Based on Deep Learning Techniques
Birlutiu, A., Burlacu, A., Kadar, M., & Onita, D. (2017) · International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC) · doi:10.1109/synasc.2017.00049 · Cited by 37
Machine learning applications to geophysical data analysis
Bougher, B. (2016) · cIRcle (University of British Columbia) · doi:10.14288/1.0308786 · Cited by 37
A Survey of Deep Q-Networks used for Reinforcement Learning: State of the Art
Hafiz, A. M. (2022) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-19-1844-5_30 · Cited by 36
Foundation models and Transformers for anomaly detection: A survey
Ammar, M. B., Mendoza, A., Belkhir, N., Manzanera, A., & Franchi, G. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103517 · Cited by 36
“The rise of deep learning Bengio (2009); LeCun et al. (2015), in particular with the introduction of Convolution Neural Networks (CNNs) LeCun et al. (2010) allowed for rapid improvement in the VAD field, with a plethora of algorithmic solutions to address the diverse challenges that may arise.”
Graph-Based Supervised Automatic Target Detection
Mishne, G., Talmon, R., & Cohen, I. (2014) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2014.2364333 · Cited by 36
“The recently introduced scattering transform, computed with a deep convolutional network [34], provides a stable translationinvariant representation and has achieved state-of-the-art results in texture classification [35], [36].”
Huang, P. C., Li, Z., Dong, Z., Han, R., Zhou, Z., Zhu, D., Liu, L., Liu, X., & Kang, J. (2019) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.9b00011 · Cited by 36
Modeling and Training of Neural Processing Systems
Kusmenko, E., Nickels, S., Pavlitskaya, S., Rumpe⋆, B., & Timmermanns, T. (2019) · ACM/IEEE International Conference on Model Driven Engineering Languages and Systems (MODELS) · doi:10.1109/models.2019.00012 · Cited by 37
“We introduce the main concepts of our neural architecture modeling language using CNNs, a class of neural networks widely-used for a variety of image processing problems [30].”
MethodAn Adversarial Approach to Private Flocking in Mobile Robot Teams
Zheng, H., Panerati, J., Beltrame, G., & Prorok, A. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.2967331 · Cited by 36
“Convolutional Neural Networks (CNNs) have proven successful in multi-class classification problem for high-dimensional input with spatial information [33].”
BackgroundA combination of feature extraction methods and deep learning for brain tumour classification
Siar, M., & Teshnehlab, M. (2021) · IET Image Processing · doi:10.1049/ipr2.12358 · Cited by 36
Eclipse Attack Detection for Blockchain Network Layer Based on Deep Feature Extraction
Dai, Q., Zhang, B., & Dong, S. (2022) · Wireless Communications and Mobile Computing · doi:10.1155/2022/1451813 · Cited by 36
“A CNN can extract layerby-layer features in a hierarchical structure, which supports mining the deep-level local complex features of samples and reconstruction of the spatial dimensions of the sample data [30].”
Background3-D Seismic Fault Detection Using Recurrent Convolutional Neural Networks With Compound Loss
Ma, X., Yao, G., Zhang, F., & Wu, D. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3275951 · Cited by 36
“By repeatedly using the convolution kernels in the input image, CNNs are capable of discovering features at arbitrary locations in the image [19].”
MethodReconstructive Sparse Code Transfer for Contour Detection and Semantic Labeling
Maire, M., Yu, S. X., & Perona, P. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-16817-3_18 · Cited by 36
Durá-Bernal, S., Wennekers, T., & Denham, S. L. (2012) · PLoS ONE · doi:10.1371/journal.pone.0048216 · Cited by 37
“However, further enforcing sparse representations, which have been identified as an essential element for similar biologically-inspired models [13,19,30,32], may improve the accurary of the model approximations.”
BackgroundMethodInfluentialMoving object recognition using multi-view three-dimensional convolutional neural networks
He, T., Mao, H., & Zhang, Y. (2016) · Neural Computing and Applications · doi:10.1007/s00521-016-2277-9 · Cited by 36
“Another advantage is that CNNs have fewer connections, which means the training stage is faster than in some other neural networks [9].”
BackgroundMultimodal Neural Network for Rapid Serial Visual Presentation Brain Computer Interface
Manor, R., Mishali, L., & Geva, A. B. (2016) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2016.00130 · Cited by 35
“Specifically, convolutional neural networks (CNNs or ConvNets) (LeCun et al., 2010) appear to be a good fit for EEG data (Cecotti and Gräser, 2011; Manor and Geva, 2015).”
BackgroundMethodInfluentialStudy on the Combined Application of CFAR and Deep Learning in Ship Detection
Wang, R., Li, J., Duan, Y., Cao, H., & Zhao, Y. (2018) · Journal of the Indian Society of Remote Sensing · doi:10.1007/s12524-018-0787-x · Cited by 36
“(5) Determine whether the validation sample error rate is in line with the user’s requirements: if not, return to step (2) and re-adjust the parameters; otherwise, the training ends, as the network optimal model has been obtained (Dahl et al. 2010).”
BackgroundA Novel Approach to Component Assembly Inspection Based on Mask R-CNN and Support Vector Machines
Huang, H., Wei, Z., & Yao, L. (2019) · Information · doi:10.3390/info10090282 · Cited by 36
“Convolutional neural networks (CNNs) show excellent performance for image processing [16,17].”
BackgroundMobahi, H., Zitnick, C. L., & Ma, Y. (2012) · 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.2012.6247869 · Cited by 36
“Another related machinery for visual recognition tasks is convolutional deep architectures [22, 23, 30, 34, 48].”
BackgroundMRAM Co-designed Processing-in-Memory CNN Accelerator for Mobile and IoT ApplicationsPreprint
Sun, B., Liu, D., Yu, L., Li, J., Liu, H., Zhang, W., & Torng, T. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.12179 · Cited by 36
“The CNN algorithm is constructed by stacking multiple computation layers for feature extraction and classification [5].”
MethodCheng, X., Chen, X., & Mallat, S. (2016) · Information and Inference A Journal of the IMA · doi:10.1093/imaiai/iaw007 · Cited by 35
SAR Target Recognition in Large Scene Images via Region-Based Convolutional Neural Networks
Cui, Z., Dang, S., Cao, Z., Wang, S., & Liu, N. (2018) · Remote Sensing · doi:10.3390/rs10050776 · Cited by 35
“The max pooling layer is to find out the largest unit in a local patch and realize a subsample on the output of the conv layer to extract local optimal features—the 2D array output of this is called the feature map [26].”
BackgroundAssessment of the Condition of Pipelines Using Convolutional Neural Networks
Vankov, Y. V., Rumyantsev, A. A., Ziganshin, S. G., Politova, T. O., Minyazev, R. S., & Рифкатович, З. А. (2020) · Energies · doi:10.3390/en13030618 · Cited by 35
“However, the reliability of the obtained results strongly depends on the choice of the type of neural network and its settings [15].”
BackgroundKayasandık, C. B., Ru, W., & Demetrio Labate, A. (2020) · Scientific Reports · doi:10.1038/s41598-020-61953-9 · Cited by 35
Kumaraswamy, B., & Poonacha, P. G. (2021) · Applied Soft Computing · doi:10.1016/j.asoc.2021.107446 · Cited by 35
Hierarchical Deep Learning Architecture for 10K Objects Classification
Katole, A. L., Yellapragada, K. P., Bedi, A. K., Kalra, S. S., & Chaitanya, M. S. (2015) · doi:10.5121/csit.2015.51408 · Cited by 35
“Deep learning based vision architectures learn to extract & represent visual features with model architectures that are composed of layers of non-linear transformations stacked on top of each other [1].”
BackgroundWerbos, P. J. (2012) · Wiley eBooks · doi:10.1002/9781118453988.ch1 · Cited by 35
Aggarwal, C. C. (2018) · Springer eBooks · doi:10.1007/978-3-319-94463-0_8 · Cited by 35
Karatzoglou, A., Schnell, N., & Beigl, M. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01418-6_7 · Cited by 35
Machine Learning Aspects of the MyShake Global Smartphone Seismic Network
Kong, Q., Inbal, A., Allen, R. M., Lv, Q., & Puder, A. (2018) · Seismological Research Letters · doi:10.1785/0220180309 · Cited by 35
“(2018), a convolutional neural network (CNN) is trained by forming images using three-component waveforms to take advantage of CNNs’ significant image processing power (LeCun et al., 2010).”
BackgroundMethodGland Segmentation in Colon Histology Images: The GlaS Challenge ContestPreprint
Sirinukunwattana, K., Pluim, J. P. W., Chen, H., Qi, X., Heng, P., Guo, Y. B., Wang, L. Y., Matuszewski, B. J., Bruni, E., Sanchez, U., et al. (2016) · Zurich Open Repository and Archive (University of Zurich) · doi:10.3929/ethz-b-000127185 · Cited by 36
Direct Maximization of Protein Identifications from Tandem Mass Spectra
Spivak, M., Weston, J., Tomazela, D. M., MacCoss, M. J., & Noble, W. S. (2011) · Molecular & Cellular Proteomics · doi:10.1074/mcp.m111.012161 · Cited by 34
“Systems that introduce learning of the features in conjunction with classification give superior results in comparison with methods that involve hand-crafted feature generation before training a classifier (15).”
ResultConvolution-based classification of audio and symbolic representations of music
Velarde, G., Cancino-Chacón, C., Meredith, D., Weyde, T., & Grachten, M. (2018) · Journal of New Music Research · doi:10.1080/09298215.2018.1458885 · Cited by 34
“The classification accuracies reported by Lidy et al. (2007) on a data-set containing classical, pop and jazz music reached 75% on symbolic representations, 86% on synthetic audio and 93% when symbolic and synthetic audio representations were used in combination.”
BackgroundMulti-scale CNN for Multi-sensor Feature Fusion in Helical Gear Fault Detection
Li, T., Zhao, Z., Sun, C., Yan, R., & Chen, X. (2020) · Procedia Manufacturing · doi:10.1016/j.promfg.2020.07.001 · Cited by 34
Model-and-Data-Driven Method for Radar Highly Maneuvering Target Detection
Wang, C., Zheng, J., Jiu, B., & Liu, H. (2021) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2021.3054073 · Cited by 34
Alaeddine, H., & Malek, J. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-05008-0 · Cited by 34
“Several techniques have been used to improve the performance of CNN: Increasing depth [31], increasing the width [19], modification of the convolution parameters [6, 32, 33] or pooling [34–42] change the activation function [27, 28, 43, 44] and reduce the number of parameters and resources.”
MethodBLSTM and CNN Stacking Architecture for Speech Emotion Recognition
Li, D., Sun, L., Xu, X., Wang, Z., Zhang, J., & Du, W. (2021) · Neural Processing Letters · doi:10.1007/s11063-021-10581-z · Cited by 34
“Pooling layer, which is inspired by the function of brain visual cortex cells [27], is also a significance operation in CNN.”
BackgroundKhayati, M. E., Elkettani, Y., & Mourchid, M. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.108288 · Cited by 34
Hybrid Deep Neural Network Model for Detection of Security Attacks in IoT Enabled Environment
Sagu, A., Gill, N. S., & Gulia, P. (2022) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2022.0130115 · Cited by 34
“The extracted features FE are provided as input to optimized CNN [43].”
MethodJing, Z., Li, P., Wu, B., Yuan, S., & Chen, Y. (2022) · Remote Sensing · doi:10.3390/rs14081950 · Cited by 34
“However, CNNs are more widely used because they avoid complex pre-processing of data and can directly input raw data relying on convolution layers to extract feature maps [25].”
MethodDeep learning solution for medical image localization and orientation detection
Zhao, Y., Zeng, K., Zhao, Y., Bhatia, P., Ranganath, M., Kozhikkavil, M. L., Li, C., & Hermosillo, G. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102529 · Cited by 34
Intelligent Monitoring System Based on Spatio–Temporal Data for Underground Space Infrastructure
Du, B., Ye, J., Zhu, H., Sun, L., & Du, Y. (2022) · Engineering · doi:10.1016/j.eng.2022.07.016 · Cited by 34
Learned Data-aware Image Representations of Line Charts for Similarity Search
Luo, Y., Zhou, Y., Tang, N., Li, G., Chai, C., & Shen, L. (2023) · Proceedings of the ACM on Management of Data · doi:10.1145/3588942 · Cited by 34
“(3) The third observation is that the methods ( i.e., Zenvisage and Qetch ) of calculating similarity based on underlying data D are better than V-CNN of computing similarity solely by image features.”
InfluentialSangher, K. S., Singh, A., Pandey, H. M., & Kumar, V. (2023) · Information · doi:10.3390/info14060349 · Cited by 34
“Therefore, we took motivation from existing crucial work and used four DL classification models based on Recurrent Neural Networks (RNNs) [87,88], Convolutional Neural Networks (CNNs) [89,90], Long ShortTerm Memory (LSTM) [91,92], and the Transformers architecture [93].”
MethodKwasigroch, A., Mikołajczyk, A., & Grochowski, M. (2017) · Advances in intelligent systems and computing · doi:10.1007/978-3-319-60699-6_81 · Cited by 34
An overview of state‐of‐the‐art image restoration in electron microscopy
Roels, J., Aelterman, J., Luong, H. Q., Lippens, S., Pižurica, A., Saeys, Y., & Philips, W. R. (2018) · Journal of Microscopy · doi:10.1111/jmi.12716 · Cited by 34
Pham, T.-A., & Hoang, V.-D. (2024) · Journal of Information and Telecommunication · doi:10.1080/24751839.2024.2317509 · Cited by 33
“This structure is demonstrated in Figure 2, known as (LeCun et al., 2010).”
ModPSO-CNN: an evolutionary convolution neural network with application to visual recognition
Tu, S., Rehman, S. U., Waqas, M. M., Rehman, O. U., Shah, Z., Yang, Z., & Koubâa, A. (2020) · Soft Computing · doi:10.1007/s00500-020-05288-7 · Cited by 33
“Y li is the i th feature map in layer l , which can be calculated as where B li is a biased matrix and K li , j is the filter, which connects the j th feature map in layer l − 1 to the i th feature map in layer l having size 2 h l 1 + 1 × 2 h l 2 + 1 (LeCun et al. 2010).”
BackgroundDatta, A., Matlock, M. K., Dang, N. L., Moulin, T. A., Woeltje, K. F., Yanik, E. L., & Swamidass, S. J. (2020) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.3033405 · Cited by 33
Image Quality Assessment of Fetal Brain MRI Using Multi‐Instance Deep Learning Methods
Largent, A., Kapse, K. J., Barnett, S. D., Asis‐Cruz, J. D., Whitehead, M. T., Murnick, J. G., Zhao, L., Andersen, N. R., Quistorff, J. L., Lopez, C., et al. (2021) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.27649 · Cited by 33
“Sujit et al. 15 used an ensemble learning of CNNs to predict image quality of T1-weighted pediatric and adult brain MRIs.”
MethodInfluentialRevisiting segmentation of lung tumors from CT images
Farheen, F., Shamil, M. S., Ibtehaz, N., & Rahman, M. S. (2022) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2022.105385 · Cited by 33
Qin, Y., Wu, Y., Li, B., Gao, S., Liu, M., & Zhan, Y. (2019) · Sensors · doi:10.3390/s19051164 · Cited by 33
“CNNs take a moving window as a filter for capturing features in image space, and the features then can be applied for classification or segmentation [19].”
BackgroundLearning across scales - A multiscale method for Convolution Neural NetworksPreprint
Haber, E., Ruthotto, L., Holtham, E., & Jun, S.-H. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.02009 · Cited by 33
“The topic has been a major field of research over the last years, after it has shown remarkable success in learning images of hand writing, natural images, videos and more (see for example [11, 10, 12] and references within).”
BackgroundIntrusion Detection System for Wireless Mesh Networks via Improved Whale Optimization
Anand, S. (2020) · Journal of Networking and Communication Systems (JNACS) · doi:10.46253/jnacs.v3i4.a2 · Cited by 32
“Particularly, a higher accuracy of 0.9985 was achieved by the presented method, which was 0.91% better than existing SVM and CNN models.”
MethodInfluentialAlnamoly, M. H., Hady, A. A., El-Kader, S. M. A., & Elhenawy, I. (2024) · IEEE Access · doi:10.1109/access.2024.3401733 · Cited by 32
“These learnable filters [24] are a smaller matrices (K) that moves through the input image (x) from left to right from top to bottom and calculates the dot product of the filter weight (w) and the corresponding input image patch.”
BackgroundUsing unsupervised learning based convolutional neural networks to solve Digital Image Correlation
Cheng, X., Ma, Q., Zhou, S., Guo, L., & Ma, S. (2024) · Optics & Laser Technology · doi:10.1016/j.optlastec.2024.111414 · Cited by 32
Face detection and recognition in an unconstrained environment for mobile visual assistive system
Chaudhry, S., & Chandra, R. (2016) · Applied Soft Computing · doi:10.1016/j.asoc.2016.12.035 · Cited by 32
Gait-Based Age Estimation with Deep Convolutional Neural Network
Zhang, S., Wang, Y., & Li, A. (2019) · International Conference on Biometrics · doi:10.1109/icb45273.2019.8987240 · Cited by 32
“Particularly, in the field of computer vision, convolutional neural network (ConvNet) [10] is a quite useful model, which can learn to extract deep feature instead of traditional handcrafted ones.”
BackgroundRevealing Fine Structures of the Retinal Receptive Field by Deep-Learning Networks
Yan, Q., Zheng, Y., Jia, S., Zhang, Y., Yu, Z., Chen, F., Tian, Y., Huang, T., & Liu, J. K. (2020) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2020.2972983 · Cited by 34
SAR image despeckling using deep CNN
Passah, A., Amitab, K., & Kandar, D. (2021) · IET Image Processing · doi:10.1049/ipr2.12104 · Cited by 32
“1) Convolution layer: This layer generates the convolution of the input x with a bank of filters W ε Rh̃×w̃×c̃×c ′ , also called kernels, and adds a bias b ε Rc ′ [41].”
Zhang, X., Li, Z., Zhang, T., Liu, F., Tang, X., Chen, P., & Jiao, L. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3106681 · Cited by 32
Metadata
- Authors
- LeCun, Y., Kavukcuoglu, K., & Farabet, C.
- Affiliations
- Courant Institute of Mathematical Sciences (US)
- ORCID
- LeCun, Yann
- Venue
- Proceedings of 2010 IEEE International Symposium on Circuits and Systems
- Publisher
- IEEE
- Published
- 2010-05
- Pages
- 253-256
- ISSN
- 0271-4302 (ISSN-L)
- OpenAlex
- W2015861736
- Semantic Scholar
- c43025c429b1fbf6f1379f61801a1b40834d62e7
- Cited by
- 2,213
- References
- 56
- Citations by year
- 2026: 30 · 2025: 112 · 2024: 168 · 2023: 201 · 2022: 223 · 2021: 242 · 2020: 255 · 2019: 240 · 2018: 206 · 2017: 183 · 2016: 142 · 2015: 94 · 2014: 59 · 2013: 21 · 2012: 21
- FWCI
- 15.63
- Type
- conference paper
- Language
- en
- Open access
- green
- hal.science
- Licence
- other-oa
- Locations
- IEEE International Symposium on Circuits and Systems proceedings (publishedVersion)
- HAL (submittedVersion, other-oa)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Keywords
- convolutional neural networks, unsupervised learning, representation learning, hierarchical feature representation, filter banks, feature pooling, nonlinear activation functions, visual object recognition, visual navigation, mobile robots, few labeled samples
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.