The DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index.
Objects365: A Large-Scale, High-Quality Dataset for Object Detection
Abstract
In this paper, we introduce a new large-scale object detection dataset, Objects365, which has 365 object categories over 600K training images. More than 10 million, high-quality bounding boxes are manually labeled through a three-step, carefully designed annotation pipeline. It is the largest object detection dataset (with full annotation) so far and establishes a more challenging benchmark for the community. Objects365 can serve as a better feature learning dataset for localization-sensitive tasks like object detection and semantic segmentation. The Objects365 pre-trained models significantly outperform ImageNet pre-trained models with 5.6 points gain (42 vs 36.4) based on the standard setting of 90K iterations on COCO benchmark. Even compared with much long training time like 540K iterations, our Objects365 pretrained model with 90K iterations still have 2.7 points gain (42 vs 39.3). Meanwhile, the finetuning time can be greatly reduced (up to 10 times) when reaching the same accuracy. Better generalization ability of Object365 has also been verified on CityPersons, VOC segmentation, and ADE tasks. The dataset as well as the pretrained-models have been released at www.objects365.org.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Existing large DETR models [3, 6, 32, 41, 44, 46] have demonstrated impressive performance on COCO test-dev [20] leaderboard.”
From DETRs Beat YOLOs on Real-time Object Detection (2024) · cites this paper as [32]
In the citing paper
“In contrast, DINO achieves the SOTA result with a publicly available SwinL [23] backbone and a public dataset Objects365 [33] (1.7M annotated images) only.”
From DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection (2022) · cites this paper as [33]
Citations
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
Self-citationFaster 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 73,472
Self-citationFaster 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
Self-citationImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Rich 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
(2016) · Cited by 31,590
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581
Research works printed but not linked: 3
- girshick, 2018, Detectron
- he, 2018, Rethinking im-agenet pre-training, arXiv preprint arXiv 1811 08883
- everingham, 0, The PASCAL Visual Object Classes Challenge 2012 (VOC2012) Results
Cited by
Sorted by citations
Items 301–400 of 677
Measuring Progress in Fine-grained Vision-and-Language Understanding
Bugliarello, E., Sartran, L., Agrawal, A., Hendricks, L. A., & Nematzadeh, A. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.87 · Cited by 23
Universal Object Detection with Large Vision Model
Lin, F., Hu, W., Wang, Y., Tian, Y., Lu, G., Chen, F., Xu, Y., & Wang, X. (2023) · International Journal of Computer Vision · doi:10.1007/s11263-023-01929-0 · Cited by 14
“Mean-while, Cai et al. (L. Cai et al., 2022) have utilized existing detection datasets (A. Gupta, Dollar, & Girshick, 2019; Kuznetsova et al., 2020; S. Shao et al., 2019) to create a large pre-training dataset through careful curation based on well-defined principles.”
Oyama, A., Hasegawa, S., Nakagawa, H., Taniguchi, A., Hagiwara, Y., & Taniguchi, T. (2023) · IEEE RO-MAN · doi:10.1109/ro-man57019.2023.10309487 · Cited by 10
“To achieve this, You Only Look Once version 5 (YOLOv5) [21] pre-trained on Ob-jects365 [22] is utilized as the object detector.”
BackgroundMethodOpenInst: A simple query-based method for open-world instance segmentation
Wang, C., Wang, G., Zhang, Q., Guo, P., Liu, W., & Wang, X. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110570 · Cited by 11
LogoMotive: Detecting Logos on Websites to Identify Online Scams - A TLD Case Study
van den Hout, T., Wabeke, T., Moura, G. C. M., & Hesselman, C. (2022) · Lecture notes in computer science · doi:10.1007/978-3-030-98785-5_1 · Cited by 10
INTERN: A New Learning Paradigm Towards General VisionPreprint
Shao, J., Chen, S., Li, Y., Wang, K., Yin, Z., He, Y., Teng, J., Sun, Q., Gao, M., Liu, J., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.08687 · Cited by 38
“More specifically, we train on the train2017 subset with ∼ 118k images, and evaluate on the val2017 set of size 5k. Objects365 [127] We use the official training and validation sets, with ∼ 608k and 30k data respectively.”
MethodInfluentialKnowledge Distillation Across Vision and Language
Fang, Z., & Yang, Y. (2023) · Studies in computational intelligence · doi:10.1007/978-3-031-32095-8_3 · Cited by 9
Integrating Multiple Models For Effective Video Retrieval and Multi-stage Search
Gia, B. T., Khanh, T. B. C., Tran, K., Trung, K. L., Doan, T. T., Le, K., Do, T., & Ngo, T. D. (2023) · International Symposium on Information and Communication Technology · doi:10.1145/3628797.3629022 · Cited by 9
“This model is first pre-trained on Objects365[18] and then fine-tuned on COCO[9], which offers a diverse range of categories and encompasses common objects found in everyday life.”
MethodIntegrating grid features and geometric coordinates for enhanced image captioning
Zhao, F., Yu, Z., Zhao, H., Wang, T., & Bai, T. (2023) · Applied Intelligence · doi:10.1007/s10489-023-05198-9 · Cited by 9
“The feature extraction model uses a combination of four datasets (COCO, Visual Genome, Open Images, and Object365 [48, 50, 51]) for training.”
Toward Open-Set Human Object Interaction Detection
Wu, M., Liu, Y., Ji, J., Sun, X., & Ji, R. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28422 · Cited by 9
“For the open-set object detector, we use the GroundingDINO (Liu et al. 2023) with Swin-B backbone which is pre-trained on COCO, O365 (Shao et al. 2019), GoldG (Kamath et al. 2021), Cap4M, OpenImage (Kuznetsova et al. 2020), ODinW-35 and RefCOCO (Kazemzadeh et al. 2014).”
MethodLarge-Scale Object Detection in the Wild With Imbalanced Data Distribution, and Multi-Labels
Pan, C., Peng, J., Bu, X., & Zhang, Z. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3421300 · Cited by 9
“of the second-largest object detection dataset Object365 [69], as shown in Table II.”
MethodCAM-Vtrans: real-time sports training utilizing multi-modal robot data
LinLin, H., Sangheang, L., & GuanTing, S. (2024) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2024.1453571 · Cited by 9
OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection
Xia, Z., Li, J., Lin, Z., Wang, X., Wang, Y., & Yang, M.-H. (2025) · neural information processing systems · doi:10.52202/085713-4084 · Cited by 18
Self-citationLeveraging Content and Context Cues for Low-Light Image Enhancement
Morawski, I., He, K., Dangi, S., & Hsu, W. H. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3543047 · Cited by 9
“For paired low-and normal-light datasets without manual annotation, we ran [76], an open-vocabulary detector, providing 365 labels from the Ob-jects365 [77] dataset, and used predictions with a confidence score over 30%.”
PoLaRIS Dataset: A Maritime Object Detection and Tracking Dataset in Pohang Canal
Choi, J., Cho, D., Lee, G., Kim, H., Yang, G., Kim, J.-W., & Cho, Y. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11128583 · Cited by 17
“‘O’, ‘G’, and ‘C’ denote pertaining using Objects365 [22], GoldG [23, 24], and Cap4M [25] dataset.”
Few-Shot Adaptation of Grounding DINO for Agricultural Domain
Singh, R., Puhl, R. B., Dhakal, K., & Sornapudi, S. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00530 · Cited by 12
“Notable among these is Grounding-DINO[33], which has been trained on very large datasets including O365[50], OI[25], GoldG[21], Cap4M[28], COCO[30], and RefC[23].”
RoboTron-Drive: All-in-One Large Multimodal Model for Autonomous Driving
Huang, Z., Feng, C., Yan, F., Xiao, B., Jie, Z., Zhong, Y., Liang, X., & Ma, L. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00751 · Cited by 49
“For the single image data, we utilize the COCO [24] and Object365 [45] datasets.”
X-Learner: Learning Cross Sources and Tasks for Universal Visual Representation
He, Y., Huang, G., Chen, S., Teng, J., Wang, K., Yin, Z., Sheng, L., Liu, Z., Qiao, Y., & Shao, J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19809-0_29 · Cited by 9
EAES: Effective Augmented Embedding Spaces for Text-Based Image Captioning
Nguyen, K. T. T. M., Bui, D. C., Trinh, T., & Vo, N. D. (2022) · IEEE Access · doi:10.1109/access.2022.3158763 · Cited by 9
“In detail, this detector is built based on the ResNeXt-152 C4 backbone and trained on a large-scale dataset including COCO [39], OpenImages [40], Objects365 [41] and Visual Genome [42].”
MethodZhang, H., Li, X., Sun, Y., Fu, X., Qiu, C., & Carroll, J. M. (2024) · IEEE Conference on Virtual Reality and 3D User Interfaces (VR) · doi:10.1109/vr58804.2024.00054 · Cited by 13
“A high-quality dataset is essential for studies on identification [71], simulation [20], design of stimuli [36], design of interactions [41] and other machine learning tasks [83].”
BackgroundYolOOD: Utilizing Object Detection Concepts for Multi-Label Out-of-Distribution Detection
Zolfi, A., Amit, G., Baras, A., Koda, S., Morikawa, I., Elovici, Y., & Shabtai, A. (2024) · 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/cvpr52733.2024.00553 · Cited by 12
“YOLO’s architecture is comprised of two components: (a) a backbone network used to extract features from the input image, and (b) three detection heads which process the image’s features at three different scales.”
BackgroundJayakrishna, S. S., & Ganesh, S. S. (2024) · Archives of Phytopathology and Plant Protection · doi:10.1080/03235408.2024.2347966 · Cited by 8
“In this article, (Li et al. 2017; Xia et al. 2018; Shao et al. 2019) the author examined the object with hardware systems to create huge datasets by extracting video to image frames, a good model requires a large number of balanced datasets. the ImageNet method was used to train the model.”
MethodZone Evaluation: Revealing Spatial Bias in Object Detection
Zheng, Z., Chen, Y., Hou, Q., Li, X., Wang, P., & Cheng, M. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3409416 · Cited by 8
“Annular zones: In practice, the centralized photographer’s bias is ubiquitous in visual datasets [25], [49], [59], [71], [77], [84].”
BackgroundElysium: Exploring Object-Level Perception in Videos via MLLM
Wang, H., Ye, Y., Wang, Y., Nie, Y., & Huang, C. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72670-5_10 · Cited by 18
Scene-Graph ViT: End-to-End Open-Vocabulary Visual Relationship Detection
Salzmann, T., Ryll, M., Bewley, A., & Minderer, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72907-2_12 · Cited by 8
Graph-Based Group Division Network for Referring Expression Comprehension
Ke, J., Wang, J., Wong, W. K., Toomey, A., & Wen, J. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3528657 · Cited by 8
“In this section, we present extensive evaluation results for the proposed method on six challenging REC benchmarks: [38], SBU [39], COCO [40], VG [41], F LICKR 30K [42], OI [42],O BJECTS 365V1 [43] AND MIMIC-CXR [44] A”
DCDL: Dual Causal Disentangled Learning for Zero-Shot Sketch-Based Image Retrieval
Li, Q., Wang, S., Zhang, W., Bai, S., Nie, W., & Liu, A.-A. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3543035 · Cited by 8
“Pre-trained vision-language models, using object locating abilities learned from auxiliary datasets [39], achieve zero-shot predictions for object detection tasks by comparing embeddings of given object proposals and texts.”
SimLTD: Simple Supervised and Semi-Supervised Long-Tailed Object Detection
Tran, P. V. (2025) · 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/cvpr52734.2025.00440 · Cited by 13
“Following prior studies, we focus on the performance gains of mAP box and AP r in our comparative analysis.”
Rostami, P., Taimori, A., Sattarzadeh, S., Shahreza, H. O., & Marvasti, F. (2020) · International Symposium on Telecommunications · doi:10.1109/ist50524.2020.9345889 · Cited by 8
“…optimized for high accuracy localization, that may either be specialized on cars as objects (e.g. Compcars [1]) or include general object categories (e.g. Objects365 [14], Openimages V4 [15], coco [12]), to effectively optimize them in terms of the localization accuracy by which they detect cars.”
BackgroundBamboo: Building Mega-Scale Vision Dataset Continually with Human–Machine Synergy
Zhang, Y., Sun, Q., Zhou, Y., He, Z., Yin, Z., Wang, K., Sheng, L., Qiao, Y., Shao, J., & Liu, Z. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02450-2 · Cited by 19
“To evaluate their effectiveness, we conduct the following experiments on Objects365 (Shao et al., 2019) dataset.”
InfluentialChinaOpen: A Dataset for Open-world Multimodal Learning
Chen, A., Wang, Z., Dong, C., Tian, K., Zhao, R., Liang, X., Kang, Z., & Li, X. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3612156 · Cited by 8
“For object recognition, we curate a set of 3,841 object labels by merging object classes from MSCOCO [24], VisualGenome [14], Objects365 [33], LVIS [8] and OpenImages [16].”
MethodUnsupervised Prompt Tuning for Text-Driven Object Detection
He, W., Chen, W., Chen, B., Yang, S., Xie, D., Lin, L., Qi, D., & Zhuang, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00250 · Cited by 9
Learning Visual Grounding from Generative Vision and Language Model
Wang, S., Kim, D., Taalimi, A., Sun, C., & Kuo, W. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00782 · Cited by 18
“Two recent datasets [26, 41] adopt image-caption pairs from web-collected datasets [4, 47] and utilizes GLIP [28] trained on Objects365 [48] and GoldG [19] to generate object boxes and uses NLP tools [15] to isolate object-level word chunks from image-level captions and expand them into referring…”
InfluentialSniffyArt: The Dataset of Smelling Persons
Zinnen, M., Hussian, A., Tran, H., Madhu, P., Maier, A., & Christlein, V. (2023) · SUMAC @ ACM Multimedia · doi:10.1145/3607542.3617357 · Cited by 8
“The availability of large-scale photographic datasets like Ima-geNet [54], COCO [41], OpenImages [38], or Objects365 [57] has enabled computer vision methods to achieve impressive performance on natural images.”
BackgroundGolroudbari, A. A., & Sabour, M. H. (2023) · doi:10.22541/au.168664884.43899660/v2 · Cited by 9
Foundation Models for Video Understanding: A SurveyPreprint
Madan, N., Møgelmose, A., Modi, R., Rawat, Y. S., & Moeslund, T. B. (2024) · doi:10.36227/techrxiv.171769139.99464428/v2 · Cited by 11
Automated Conversion of Music Videos into Lyric Videos
Ma, J., Rao, A., Wei, L., Kazi, R. H., Shin, H. V., & Agrawala, M. (2023) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3586183.3606757 · Cited by 7
Research on Deep Learning-Based Vehicle and Pedestrian Object Detection Algorithms
Zhang, X., Huang, H., Yang, J., & Jiang, S. (2024) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlviii-4-w10-2024-213-2024 · Cited by 7
“Publicly available datasets include PASCAL VOC, ImageNet, MSCOCO, and Objects365 (Everingham et al., 2010; Deng et al., 2009; Lin et al., 2014; Shao et al., 2019).”
MethodSelf-citationA Survey of Multimodal Hallucination Evaluation and Detection
Chen, Z., Min, Y., Zhang, J., Yan, B., Wang, J., Wang, X., & Shan, S. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02756-9 · Cited by 31
Multi-Task Learning with Task-Specific Feature Filtering in Low-Data Condition
Lee, S.-W., Lee, R., Seo, M., Park, J., Noh, H., Ju, J., Jang, R.-Y., Lee, G., Choi, M.-S., & Choi, D. (2021) · Electronics · doi:10.3390/electronics10212691 · Cited by 7
“Convolutional neural networks (CNNs) are nowadays the state-of-the-art methods for a wide range of computer vision tasks, thanks to the large-scale public datasets [1–3] and high performance accelerators like graphical processing units (GPUs).”
BackgroundFewSOL: A Dataset for Few-Shot Object Learning in Robotic Environments
Jishnu Jaykumar, P., Chao, Y.-W., & Yu, X. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10161143 · Cited by 8
“While large-scale datasets collected from the Internet such as ImageNet [4], Visual Genome [5], Objects365 [6], and open images [7] contain large numbers of object categories, these datasets are useful for learning visual representations but are not very suitable to learn object representations for robot manipulation…”
BackgroundClipCrop: Conditioned Cropping Driven by Vision-Language Model
Zhong, Z., Cheng, M., Wu, Z., Yuan, Y., Zheng, Y., Li, J., Han, H., Lin, S. C.-F., Sato, Y., & Sato, I. (2023) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw60793.2023.00037 · Cited by 10
“This stage is trained using publicly available detection datasets including OpenImages-V 4 [24], Objects 365 (O365) [53], and Visual Genome [22].”
MethodSelf-citationLiu, J., Niu, J., Li, W., Li, X., He, B., Zhou, H., Liu, Y., Ding, L., Wang, B., & Zhang, W. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3405998 · Cited by 7
“Generic computer vision datasets consist of various types of data, including image classification datasets [50], [51], action recognition datasets [52], [53], [54], [55], object detection and tracking datasets [56], [57], [58], [59], and semantic segmentation datasets [56], [57], [60].”
BackgroundLiu, Y. F., Xiao, L., Shao, W., Peng, L. F., & Liu, Q. H. (2024) · IEEE Antennas and Wireless Propagation Letters · doi:10.1109/lawp.2024.3456838 · Cited by 7
Generating Enhanced Negatives for Training Language-Based Object Detectors
Zhao, S., Zhao, L., G, V. K. B., Suh, Y., Metaxas, D., Chandraker, M. K., & Schulter, S. (2024) · 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/cvpr52733.2024.01290 · Cited by 7
“Detection datasets with a fixed label space provide negative classes implicitly or explicitly, with exhaustive [23, 42] or federated [6, 16] annotations, respectively.”
BackgroundMethodInfluentialAnnotation Quality Versus Quantity for Object Detection and Instance Segmentation
Agnew, C., Scanlan, A., Denny, P., Grua, E. M., van de Ven, P. W. J., & Eising, C. (2024) · IEEE Access · doi:10.1109/access.2024.3467008 · Cited by 10
“…[2] published in 2012, 164,000 images for the Common Objects in Context (COCO) dataset [3] in 2014 and finally 2 million images for The associate editor coordinating the review of this manuscript and approving it for publication was Andrea Bottino . the Objects365 dataset released in 2019 [4].”
BackgroundSketchQL: Video Moment Querying with a Visual Query Interface
Wu, R., Chunduri, P., Payani, A., Chu, X., Arulraj, J. J. P., & Rong, K. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3677140 · Cited by 7
“In this section, we present an overview of the key components and workflow of SketchQL.”
ControlCap: Controllable Region-Level Captioning
Zhao, Y., Liu, Y., Guo, Z., Wu, W., Gong, C., Ye, Q., & Wan, F. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72920-1_2 · Cited by 15
“We evaluate the controllable ability using three vision tasks, including object localization on ImageNet-1K [10], object detection on Object365 [41], and scene text spotting on ICDAR2015 [23].”
MethodEbaid, D. B., Madbouly, M. M., & El-Zoghabi, A. A. (2023) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-023-00260-3 · Cited by 9
“In addition, we exclude the VL_PTM Faster R-CNN BERT Fusion encoder (single stream) OSCAR [107] Faster R-CNN BERT Fusion encoder (single stream) ViLBERT [105] Faster R-CNN BERT Fusion encoder (dual stream) LXMERT [108] Faster R-CNN BERT Fusion encoder (dual stream) ALBEF [109] ViT [118] BERT Fusion encoder (dual strea…”
Robust Geometry-Preserving Depth Estimation Using Differentiable Rendering
Zhang, C., Yin, W., Yu, G., Wang, Z., Chen, T., Fu, B., Zhou, J. T., & Shen, C. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00822 · Cited by 7
“in sharp contrast to many other vision tasks, such as image segmentation [17, 18, 41, 16] and object detection [16, 30], which consistently benefit from the increasing number of diverse training data.”
ResultSelf-citationA Crowdsourcing Repeated Annotations System for Visual Object Detection
Hu, Y., Ou, Z., Xu, X., & Song, M. (2019) · International Conference on Vision, Image and Signal Processing (ICVISP) · doi:10.1145/3387168.3387242 · Cited by 7
“To learn a good detector, it is necessary to have a large number of training images with ground truth annotations [18].”
BackgroundAffordance Transfer Learning for Human-Object Interaction DetectionPreprint
Hou, Z., Yu, B., Qiao, Y., Peng, X., & Tao, D. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.02867 · Cited by 7
“In detail, we select objects that are labeled as COCO classes from Object365 validation dataset to evaluate the affordance recognition of objects on new domain.”
BackgroundMethodInfluentialTowards Open Vocabulary Object Detection without Human-provided Bounding BoxesPreprint
Gao, M., Chen, X., Niebles, J. C., Li, J., Xu, R., Liu, W., & Xiong, C. (2021) · arXiv (Cornell University) · Cited by 7
Self-citationSteinert, P., Wagenpfeil, S., Frommholz, I., & Hemmje, M. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681711 · Cited by 8
Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models
Jocher, G., Qiu, J., Liu, M., Lyu, S., Akyon, F. C., & Kalfaoglu, M. E. (2026) · arXiv · doi:10.48550/arxiv.2606.03748 · Cited by 16
“For the reported COCO detection benchmarks, we use a two-stage training recipe: all model sizes are first pretrained on Objects365-v1 [63] for 150 epochs, and are then fine-tuned on COCO.”
Vision-based power line cables and pylons detection for low flying aircraft
Gwizdała, J., Öner, D., Roy, S. K., Shah, M. A., Eberhard, A., Egorov, I., Krüsi, P., Yakushev, G., & Fua, P. (2025) · Machine Vision and Applications · doi:10.1007/s00138-025-01664-1 · Cited by 6
“We choose these threshold values for each data fold separately, maximizing their respective validation performance, therefore the reported mean metrics represent the best achievable performance, providing very strong baseline results for our models to compare with.”
Optimizing Class Imbalance in Facial Expression Recognition Using Dynamic Intra-Class Clustering
Li, Q., Fu, K., Liu, J., Li, Y., Ren, Q., Xu, K., Fu, J., Liu, N., & Yuan, Y. (2025) · Biomimetics · doi:10.3390/biomimetics10050296 · Cited by 6
“Deep neural networks (DNNs) [1] have demonstrated remarkable advancements in vision tasks such as image classification [2], object detection [3,4], and semantic segmentation [5,6], with their success heavily reliant on large-scale, high-quality datasets [7,8].”
A Hierarchical Semantic Distillation Framework for Open-Vocabulary Object Detection
Fu, S., Yan, J., Yang, Q., Wei, X., Xie, X., & Zheng, W. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3607729 · Cited by 10
“We follow previous works to conduct experiments on OV-LVIS [56] and OV-COCO [57] datasets and test the transfer ability of the model trained on OV-LVIS to COCO [57] and Objects365 [58] dataset.”
Practical Video Object Detection via Feature Selection and Aggregation
Shi, Y., Zhang, T., & Guo, X. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-025-02700-3 · Cited by 11
“Thanks to the development of hardware, large-scale datasets [34,30,47] and sophisticated network structures [50,24,62,59,28,26,40,53,39,10], the performance of object detection has continuously improved.”
Self-citationDual-Rate Dynamic Teacher for Source-Free Domain Adaptive Object Detection
He, Q., Wu, X., He, J.-Y., & Li, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00200 · Cited by 6
Improving Human-Object Interaction Detection via Virtual Image Learning
Fang, S., Liu, S., Li, J., Jiang, G., Lin, X., & Ji, R. (2023) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3581783.3611735 · Cited by 6
“Instead of using latent features which lack visual representations, ATL [12] directly extends the original dataset with additional datasets [21, 30].”
BackgroundSelf-citationLearn How to See: Collaborative Embodied Learning for Object Detection and Camera Adjusting
Shen, L., Huo, C., Xu, N., Han, C., & Wang, Z. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i5.28281 · Cited by 7
“In recent years, due to the increased training data (Lin et al. 2014; Shao et al. 2019; Kuznetsova et al. 2020) and GPU capability, the performance of object detectors has significantly improved.”
BackgroundPredictive and Near-Optimal Sampling for View Materialization in Video Databases
Xu, Y., Zhang, D., Zhang, S., Wu, S., Feng, Z., & Chen, G. (2024) · Proceedings of the ACM on Management of Data · doi:10.1145/3639274 · Cited by 7
Benchmarking Out-of-Distribution Detection in Visual Question Answering
Shi, X., & Lee, S. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00540 · Cited by 6
“We use a VinVL-based object detection model [55] pretrained on COCO [29], OpenImages [24], Ob-jects365 [39], and Visual Genome [25] rather than the original BUTD features.”
MethodUniDetector: Towards Universal Object Detection With Heterogeneous Supervision
Wang, Z., Li, Y., Chen, X., Lim, S.-N., Torralba, A. B., Zhao, H., & Wang, S. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3411595 · Cited by 6
“Datasets: To simulate images of multiple sources and heterogeneous label spaces, we adopt three popular object detection datasets for training: COCO [6], Objects365 [91], and Open-Images [92].”
Enhancing Vision-Language Pre-Training with Rich Supervisions
Gao, Y., Shi, K., Zhu, P., Belval, E., Nuriel, O., Appalaraju, S., Ghadar, S., Tu, Z., Mahadevan, V., & Soatto, S. (2024) · 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/cvpr52733.2024.01280 · Cited by 7
“…many VLMs [10, 56] use a mixture of supervised pre-training tasks (e.g. object detection, VQA, cap-tioning, etc), relying on human manually labeled datasets such as COCO [39], Object365 [51], VQA [20], etc as well as datasets generated in automated fashion such as LAION-5B [50], WebLi-10B [10].”
MethodWang, X., Ren, W., Chen, X., Fan, H., Tang, Y., & Han, Z. J. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681212 · Cited by 6
“Uni-YOLO is trained on various detection datasets, including Ob-ject365 [39], Pascal VOC [9], COCO [24] and OpenImages [16].”
Rethinking Open-Set Object Detection: Issues, A New Formulation, and Taxonomy
Hosoya, Y., Suganuma, M., & Okatani, T. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02479-3 · Cited by 8
“Specifically, we apply normalized graph cut (Shi and Malik, 2000) to obtain three clusters, ensuring any pairs of the clusters share the minimum co-occurrence.”
Kajian Pengaruh Dataset dan Bias Dataset terhadap Performa Akurasi Deteksi Objek
Tiyar, R. I., & Fudholi, D. H. (2021) · Petir · doi:10.33322/petir.v14i2.1350 · Cited by 6
“Teknik Pengumpulan Data Teknik Anotasi [3] Data dikumpulkan melalui Flicker agar memiliki variasi yang banyak.”
BackgroundSpatial likelihood voting with self-knowledge distillation for weakly supervised object detection
Chen, Z., Fu, Z., Huang, J., Tao, M., Jiang, R., Tian, X., Chen, Y., & Hua, X. (2021) · Image and Vision Computing · doi:10.1016/j.imavis.2021.104314 · Cited by 7
Environment-Adaptive Object Detection Framework for Autonomous Mobile Robots
Shin, D., Cho, J., & Kim, J. (2022) · Sensors · doi:10.3390/s22197647 · Cited by 6
“The classifier classified the Objects365 data into environment-specific datasets.”
MethodInfluentialResearch Progress on Vision–Language Multimodal Pretraining Model Technology
Wang, H., Huang, R., & Zhang, J. (2022) · Electronics · doi:10.3390/electronics11213556 · Cited by 6
Recent advances in deep learning based computer vision
Ma, D. (2022) · International Conference on Computers, Information Processing and Advanced Education (CIPAE) · doi:10.1109/cipae55637.2022.00044 · Cited by 6
“Picture segmentation separates an image into many sections based on gray scale, color, spatial texture, geometry, and other parameters, so that these characteristics are consistent or comparable in one part but visibly different in another [14].”
BackgroundYao, Y., Liu, P., Zhao, T., Zhang, Q., Liao, J., Fang, C., Lee, K., & Wang, Q. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i7.28485 · Cited by 7
Integration of Metaverse Recordings in Multimedia Information Retrieval
Steinert, P., Wagenpfeil, S., Frommholz, I., & Hemmje, M. (2024) · International Conference on Software and Computer Applications (ICSCA) · doi:10.1145/3651781.3651802 · Cited by 6
GiT: Towards Generalist Vision Transformer Through Universal Language Interface
Wang, H., Tang, H., Li, J., Shi, S., Naeem, M. F., Li, H., Schiele, B., & Wang, L. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73397-0_4 · Cited by 12
Self-citationDomain-Aware Dynamic NetworksPreprint
Zhang, T., Wu, B., Wang, X. Y., Gonzalez, J. E., & Keutzer, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.13237 · Cited by 6
“domains[45, 27] and use that dataset to train a single gigantic neural network (Figure 2).”
BackgroundSelf-citationJoint COCO and Mapillary Workshop at ICCV 2019: COCO Instance Segmentation Challenge TrackPreprint
Li, Z., Ma, Y., Chen, Y., Zhang, X., & Sun, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.02475 · Cited by 6
“In the final submission, we leverage Objects365 detection datasets [12] for pre-training, which includes 365 object categories, and 600K training/30K validation images.”
MethodSelf-citationClosing the Generalization Gap in One-Shot Object DetectionPreprint
Michaelis, C., Bethge, M., & Ecker, A. S. (2021) · arXiv (Cornell University) · Cited by 6
Language-Conditioned Detection Transformer
Cho, J. H., & Krähenbühl, P. (2024) · 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/cvpr52733.2024.01570 · Cited by 6
“We consider three primary benchmarks to evaluate our final model ( DECOLA Phase 2): open-vocabulary LVIS [19], standard LVIS [20], and direct zero-shot evaluation to LVIS, COCO [36], Object365 [53], and OpenImages [32].”
MethodProduction-based progress monitoring of rebar tying using few-shot learning and kernel density
Gao, B., Yang, B., Xiao, H., & Zhou, Y. (2025) · Alexandria Engineering Journal · doi:10.1016/j.aej.2024.12.098 · Cited by 5
Tadjine, C., Ouafi, A., Ahmed, A. T., Hillali, Y. E., & Rivenq, A. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01673-3 · Cited by 5
“Additionally, there are more available datasets for object detection, such as ImageNet [22], PASCAL Visual Object Classes (PASCAL VOC) [23], Objects 365 [24] and Open Images Dataset V4 [25].”
DreamLayer: Simultaneous Multi-Layer Generation via Diffusion Model
Huang, J., Yan, P., Cai, J., Liu, J., Wang, Z., Wang, Y., Wu, X., & Li, G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00321 · Cited by 16
“If the base prompt lacks sufficient foreground objects, GPT-4 selects a suitable category from the Object365 [29] dataset.”
Computational approaches for multi-model object detection using deep learning
Waghale, P., Talekar, R., Mungale, N., Damahe, L. B., Mungale, S. G., & Mungale, S. (2024) · AIP conference proceedings · doi:10.1063/5.0224652 · Cited by 5
PartGLEE: A Foundation Model for Recognizing and Parsing Any Objects
Li, J., Wu, J., Zhao, W., Bai, S., & Bai, X. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73226-3_27 · Cited by 8
Self-citationGLSD: a global large-scale ship database with baseline evaluations
Shao, Z., Wang, Y., Wang, J., Deng, L., Huang, X., Lü, T., Luo, F., & Zhang, R. (2024) · Geo-spatial Information Science · doi:10.1080/10095020.2024.2416896 · Cited by 6
“From early datasets, like COCO [11], to the recent benchmarks, like Objects365 [22], large-scale datasets have always been preferred choices by deep learning algorithm developers, as they play an essential role in evaluating the performance of object classification and detection tasks.”
Moutik, O., Tigani, S., Saadane, R., & Chehri, A. (2021) · Procedia Computer Science · doi:10.1016/j.procs.2021.09.287 · Cited by 5
A Technical Approach for Promotion of Yakshagana - A Literature Survey
Murthy, A. K., & Nethravathi, P. S. R. (2021) · International Journal of Case Studies in Business IT and Education · doi:10.47992/ijcsbe.2581.6942.0132 · Cited by 7
“, et al (2019) [70], For object detection and semantic segmentation, Objects365 can be a better feature learning dataset.”
BackgroundMulti-attribute object detection benchmark for smart city
Wang, Y., Yang, Z., Liu, R., Deng, L., Lai, Y., Ouyang, L., Fang, L., & Han, Y. (2022) · Multimedia Systems · doi:10.1007/s00530-022-00971-1 · Cited by 5
“In many scenarios of smart city governance, detection algorithms [6, 9, 11, 13, 14, 17, 19, 20, 22, 23] face great challenges brought by object shape and size changes, orientation changes, illumination changes, weather changes, occlusion, blur, aggregation, background interference, etc.”
BackgroundA Survey of Full-Cycle Cross-Modal Retrieval: From a Representation Learning Perspective
Wang, S., Zhu, L., Shi, L., Mo, H., & Tan, S. (2023) · Applied Sciences · doi:10.3390/app13074571 · Cited by 5
Localizing Active Objects from Egocentric Vision with Symbolic World Knowledge
Wu, T.-L., Zhou, Y., & Peng, N. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.304 · Cited by 6
Smelly, dense, and spreaded: The Object Detection for Olfactory References (ODOR) dataset
Zinnen, M., Madhu, P., Leemans, I., Bell, P., Hussian, A., Tran, H., Hürriyetoğlu, A., Maier, A., & Christlein, V. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124576 · Cited by 6
“However, these datasets still fall short when compared to real-world counterparts like OpenImages (with 600 categories) or Objects365 [56] (with 365 categories) in terms of fine-grained categorization.”
Luong, T. V., Nguyen, H. S. P., Dinh, D. K., Duong, V. H., Vo, D. H. S., Vu, H., Hoang, M. T., & Nguyen, T. C. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00696 · Cited by 5
“Finally, the Co-DETR model [38] is initialized with pre-trained weights from the Object 365 dataset [27] and then fine-tuned on Data V1 and V2, employing a learning rate of 1e-4.”
MethodMPCHAT: Towards Multimodal Persona-Grounded Conversation
Ahn, J., Song, Y., Yun, S., & Kim, G. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.189 · Cited by 21
D-ViSA: A Dataset for Detecting Visual Sentiment from Art Images
Kim, S., An, C., Cha, J., Kim, D., & Park, E. (2023) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw60793.2023.00328 · Cited by 5
“To validate the categorical labels, we followed the similar approach done on several prior works on image classification tasks [58], by recruiting three inspectors, who had extensive experience in data annotations and labeling procedures.”
MethodRethinking Training from Scratch for Object DetectionPreprint
Li, Y., Zhang, H., & Zhang, Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.03112 · Cited by 5
“And with more and more massive scale detection datasets like Open Images [40], Objects365 [41] recently introduced, we need to develope more efficient training strategies to tackle these cases.”
BackgroundMulti-Task Self-Training for Learning General RepresentationsPreprint
Ghiasi, G., Zoph, B., Cubuk, E. D., Le, Q. V., & Lin, T.-Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.11353 · Cited by 5
“showed similar findings using the Objects365 dataset [49].”
BackgroundResultInfluentialMultimodal Inplace Prompt Tuning for Open-set Object Detection
Li, G., Zhang, M., Zheng, X., Chen, P., Wang, Z., Shen, Y., Zhuge, M., Wu, C., Chao, F., Li, K., et al. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681275 · Cited by 6
“For each category, five random images were chosen and processed through the vision encoder and ROIPool [46] to extract specific regional features [59].”
InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
Wang, W., Chen, Z., Liu, Y., Cao, Y., Wang, W., Zhu, X., Lu, L., Lü, T., Qiao, Y., & Dai, Y. (2025) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-031-94969-2_2 · Cited by 4
Spatial Preference Rewarding for MLLMs Spatial Understanding
Qiu, H., Gao, P., Lu, L., Zhang, X., Shao, L., & Lu, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00075 · Cited by 4
“To construct preference data, we randomly select 10k images with object annotations from the training set of Objects365 Dataset [46], then construct random regions to query models to generate grounded region descriptions.”
Self-citation
Metadata
- Authors
- Shao, S., Li, Z., Zhang, T., Peng, C., Yu, G., Zhang, X., Li, J., & Sun, J.
- Affiliations
- Megvii (China) (CN)
- Venue
- 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
- Publisher
- IEEE
- Published
- 2019-10
- Pages
- 8429-8438
- OpenAlex
- W2983943451
- Cited by
- 673
- References
- 42
- Citations by year
- 2026: 70 · 2025: 155 · 2024: 180 · 2023: 132 · 2022: 67 · 2021: 50 · 2020: 17 · 2019: 2
- FWCI
- 11.08
- Type
- conference paper
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- 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 › Visual Attention and Saliency Detection
- Keywords
- bounding box annotation, annotation pipeline, pre-trained models, representation learning, semantic segmentation, benchmark, COCO benchmark, CityPersons, ADE20K
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
