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
Flagged first, then by citations
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 201–300 of 677
COCO-O: A Benchmark for Object Detectors under Natural Distribution Shifts
Mao, X., Chen, Y., Zhu, Y. P., Chen, D., Su, H., Zhang, R., & Xue, H. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00583 · Cited by 32
“scratch without any pre-training data; 2) using pre-trained checkpoint to initialize the backbone; 3) leveraging datasets like Object365 [71] to pre-train the overall detector and then fine-tuning.”
MethodInfluentialSiamese-DETR for Generic Multi-Object Tracking
Liu, Q., Li, Y., Jiang, Y., & Fu, Y. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3416880 · Cited by 22
“For example, GLIP-T (B) [43] utilizes the pretrained BERT [10] to extract text features and the detection model is trained on Objects365 [46].”
MethodLP-OVOD: Open-Vocabulary Object Detection by Linear Probing
Pham, C., Vu, T., & Nguyen, K. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00083 · Cited by 20
“We evaluate the transfer learning performance of our approach on Objects365 [33] and PASCAL VOC [6] following the protocol in [5,10].”
MethodDeCLIP: Decoupled Learning for Open-Vocabulary Dense Perception
Wang, J., Chen, B., Li, Y., Kang, B., Chen, Y., & Tian, Z. (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/cvpr52734.2025.01381 · Cited by 19
UnityShip: A Large-Scale Synthetic Dataset for Ship Recognition in Aerial Images
He, B., Li, X., Huang, B., Gu, E., Guo, W., & Wu, L. (2021) · Remote Sensing · doi:10.3390/rs13244999 · Cited by 19
Deeply Unsupervised Patch Re-Identification for Pre-Training Object Detectors
Ding, J., Xie, E., Xu, H., Jiang, C., Li, Z., Luo, P., & Xia, G.-S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3164911 · Cited by 38
“One solution to eliminate the misalignment [7] is to pre-train representations directly on a largescale and high-quality object detection dataset; but annotations of such datasets are time-consuming, laborious, and even hard to obtain for some areas.”
BackgroundMethodSelf-citationKaul, P., Li, Z., Yang, H., Dukler, Y., Swaminathan, A. D., Taylor, C. J., & 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/cvpr52733.2024.02571 · Cited by 37
“Many LVLMs train on COCO directly or indirectly, thus to demonstrate generality we apply THRONE to the Ob-jects365 dataset [42].”
InfluentialSelf-citationLi, J., Lammers, K., Yin, X., Yin, X., He, L., Sheng, J., Lu, R., & Lu, R. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.109908 · Cited by 22
Self-citationLaVin-DiT: Large Vision Diffusion Transformer
Wang, Z., Xia, X., Chen, R., Yu, D., Wang, C., Gong, M., & Liu, T. (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/cvpr52734.2025.01868 · Cited by 35
“This dataset comprises approximately 3.2 million unique images [15, 17, 33, 48, 71] and 0.6 million unique videos [24, 27, 50], covering over 20 tasks: • Image-based tasks : object detection, instance segmentation, panoptic segmentation, pose estimation, edge extraction, depth estimation, surface…”
InfluentialLong-tail Detection with Effective Class-Margins
Cho, J. H., & Krähenbühl, P. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20074-8_40 · Cited by 18
“This is reflected in the common evaluation metric: meanaverage-precision (mAP) [11,15,21,26,37].”
BackgroundDe-biased Teacher: Rethinking IoU Matching for Semi-supervised Object Detection
Wang, K., Zhuang, J., Li, G., Fang, C., Cheng, L., Lin, L., & Zhou, F. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i2.25355 · Cited by 18
“COCO-val2017 and Object365-val are both used to evaluate the approach.”
InfluentialCOCONut: Modernizing COCO Segmentation
Deng, X., Yu, Q., Wang, P., Shen, X., & Chen, L.-C. (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/cvpr52733.2024.02065 · Cited by 20
“Specifically, DETA [44] (utilizing a Swin-L backbone [39] trained with Objects365 [54] and COCO detection set [36]) and kMaX-DeepLab [68] (featuring a ConvNeXt-L backbone [40] trained with COCO panoptic set [30]) are utilized to generate box and mask proposals for ‘thing’ and ‘stuff’, respectively.”
BackgroundMethodInfluentialEfficient video annotation with visual interpolation and frame selection guidance
Kuznetsova, A., Talati, A., Luo, Y., Simmons, K., & Ferrari, V. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00311 · Cited by 19
“Recently several large scale image datasets have appeared [19, 35, 9], as well as large-scale tracking benchmarks [13, 5], but they required tremendous annotation resources to create [19, 41].”
BackgroundFoundation 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/v1 · Cited by 24
RescueADI: Adaptive Disaster Interpretation in Remote Sensing Images With Autonomous Agents
Liu, Z., Zhao, D., Yuan, B., & Jiang, Z. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3532594 · Cited by 23
“Unlike object classification datasets of natural images [31], [32], [33], [34] that focus on the foreground object in the image, scene classification in RSIs takes into account the background information.”
Visual Grounding in 2D and 3D: A unified perspective and survey
Guo, K., Huang, Y., Jia, T., Song, X., Sun, S., Wei, H., Han, X.-F., Huang, S., Strisciuglio, N., & Li, S. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103625 · Cited by 19
Learning Semantic Segmentation from Multiple Datasets with Label Shifts
Kim, D.-W., Tsai, Y., Suh, Y., Faraki, M., Garg, S., Chandraker, M. K., & Han, B. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19815-1_2 · Cited by 18
“Also, a recent concurrent work [51] replaces class labels with text descriptions for achieving multi-dataset training, where they add open datasets [2, 40] to training.”
BackgroundWang, J., Xu, S., Dai, Y., & Gao, S. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3223878 · Cited by 17
“The dataset, including images of lamps, air conditioners, and televisions (TV), is collected from the Objects365 dataset [66] and the Common Objects in Context (COCO) dataset [67].”
MethodFew-shot Adaptive Object Detection with Cross-Domain CutMix
Nakamura, Y., Ishii, Y., Maruyama, Y., & Yamashita, T. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-26351-4_45 · Cited by 17
Cascade Semantic Prompt Alignment Network for Image Captioning
Li, J., Zhang, L., Zhang, K., Hu, B., Xie, H., & Mao, Z. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3343520 · Cited by 18
“The object detector employed a composite dataset consisting of four distinct sources, namely Visual Genome [68], Open Images [69], and Object365 [70] and MSCOCO.”
MethodSelf-citationDETR Does Not Need Multi-Scale or Locality Design
Lin, Y., Yuan, Y., Zhang, Z., Li, C., Zheng, N., & Hu, H. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00602 · Cited by 20
“Further introducing the Objects365 [40] as the pre-training dataset, our method reaches 63 .”
MethodImproving CLIP Fine-tuning Performance
Wei, Y., Hu, H., Xie, Z., Liu, Z., Zhang, Z., Cao, Y., Bao, J., Chen, D., & Guo, B. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00501 · Cited by 20
“The pre-trained weights are applied to numerous down-stream tasks including image segmentation [66, 16], object detection [36, 46] and video recognition [24, 15].”
MethodBai, S., Cai, Y., Chen, R., Chen, K., Chen, X., Cheng, Z., Deng, L., Ding, W., Gao, C., Ge, C., et al. (2025) · arXiv · doi:10.48550/arxiv.2511.21631 · Cited by 27
“We begin by aggregating widely used open-source datasets, including COCO (Lin et al., 2014), Objects365 (Shao et al., 2019), OpenImages (Kuznetsova et al., 2020), and RefCOCO/+/g (Kazemzadeh et al., 2014; Mao et al., 2016).”
Self-citationRevisiting Referring Expression Comprehension Evaluation in the Era of Large Multimodal Models
Chen, J., Wei, F., Zhao, J., Song, S., Wu, B., Peng, Z., Chan, S.-H. G., & Zhang, H. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00056 · Cited by 48
“Unlike object detection [4, 22, 28, 50, 53], which operates within fixed categories and a single visual modality, REC necessitates understanding free-form text to locate objects of any category.”
GAIA-Universe: Everything is Super-Netify
Peng, J., Chang, Q., Yin, H., Bu, X., Sun, J., Xie, L., Zhang, X., Tian, Q. C., & Zhang, Z. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3276392 · Cited by 16
“With GAIA, we achieve promising results on COCO, Objects365, Open Images, BDD100 k, and UODB which is a collection of datasets including KITTI, VOC, WiderFace, DOTA, Clipart, Comic, and more.”
BackgroundInfluentialSelf-citationArchangel: A Hybrid UAV-Based Human Detection Benchmark With Position and Pose Metadata
Shen, Y., Lee, Y., Kwon, H., Conover, D., Bhattacharyya, S. S., Vale, N., Gray, J. D., Leong, G. J., Evensen, K., & Skirlo, F. E. (2023) · IEEE Access · doi:10.1109/access.2023.3299235 · Cited by 16
“5° various Mannequin 1920×1080 Mannequins, vehicles [15-50] increment by 5 [15-50] increment by 5 120° 45° Real 1304×978 Real people [15-50] increment by 5 [20-50] increment by 5 45° 22.”
BackgroundMethodDistilling DETR with Visual-Linguistic Knowledge for Open-Vocabulary Object Detection
Li, L., Miao, J., Shi, D., Tan, W., Ren, Y., Yang, Y., & Pu, S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00598 · Cited by 20
“…ability [37] of the open-vocabulary object detection model, following ViLD [11], we directly evaluate the LVIS-trained model on COCO, Objects365 [31] and Pascal VOC [6] datasets by just replacing the classifier with text embed-dings, which is produced by feeding categories names of those…”
MethodGenerative Region-Language Pretraining for Open-Ended Object Detection
Lin, C., Jiang, Y., Qu, L., Yuan, Z., & Cai, J. (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/cvpr52733.2024.01324 · Cited by 23
“Our model is trained on 16 A100 GPUs.”
BackgroundMethodDisentangled Pre-Training for Human-Object Interaction Detection
Li, Z., Li, X., Ding, C., & Xu, X. (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/cvpr52733.2024.02663 · Cited by 16
Self-citationSpatial-temporal initialization dilemma: towards realistic visual tracking
Liu, C., Yuan, Y., Chen, X. Y., Lu, H., & Wang, D. (2024) · Visual Intelligence · doi:10.1007/s44267-024-00068-5 · Cited by 16
“Among all the detectors, EVA [56] and DETA [57] are additionally pretrained with the large-scale, high-quality object detection dataset Objects365 [58].”
Zero‐shot framework for construction equipment task monitoring
Jeoung, J., Jung, S., & Hong, T. (2025) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.13506 · Cited by 15
“This approach is consistent with the training data, as datasets such as Object365 (Shao et al., 2019), which are The tracking module tracks the detected construction equipment across the sequential frames of the clip, generating unique IDs and bounding boxes to continuously monitor the location of…”
CineMaster: A 3D-Aware and Controllable Framework for Cinematic Text-to-Video Generation
Wang, Q., Luo, Y., Shi, X., Jia, X., Lu, H., Xue, T., Wang, X., Wan, P., Zhang, D., & Gai, K. (2025) · ACM SIGGRAPH Conference (SIGGRAPH) · doi:10.1145/3721238.3730755 · Cited by 34
“The images are collected from COCO [Lin et al. 2014] and Object365 [Shao et al. 2019] which provide more categories and instance segmentation annotations.”
Discovering Human-Object Interaction Concepts via Self-Compositional Learning
Hou, Z., Yu, B., & Tao, D. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19812-0_27 · Cited by 15
Zhang, Z., Zhang, Q., Hu, X., Zhang, M., & Zhu, D. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.05.026 · Cited by 15
A Framework to Estimate the Key Point Within an Object Based on a Deep Learning Object Detection
Kurdthongmee, W., Suwannarat, K., & Wattanapanich, C. (2023) · HighTech and Innovation Journal · doi:10.28991/hij-2023-04-01-08 · Cited by 15
“Prior research has emphasized improving the scale, resolution, and diversity of the training data [3, 11-19].”
MethodHigh-Quality Object Detection Method for UAV Images Based on Improved DINO and Masked Image Modeling
Lu, W., Niu, C., Lan, C., Liu, W., Wang, S., Yu, J., & Hu, T. (2023) · Remote Sensing · doi:10.3390/rs15194740 · Cited by 16
Zhang, X., Zeng, Y., Zhang, J., & Li, H. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.40 · Cited by 21
Self-citationUnified Visual Relationship Detection with Vision and Language Models
Zhao, L., Yuan, L., Gong, B., Cui, Y., Schroff, F., Yang, M., Adam, H., & Liu, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00641 · Cited by 21
“However, allowing fine-tuning these modules leads to performance boosts when we train unified models Object detection datasets (12) w/o Objects365 [62] for object detection − 3 .”
BackgroundMethodInfluentialSemi-supervised Open-World Object Detection
Mullappilly, S. S., Gehlot, A. S., Anwer, R. M., Khan, F. S., & Cholakkal, H. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i5.28227 · Cited by 15
“6 compares our SS-OWFormer with OW-DETR on the Objects365 [25] benchmark.”
MethodMamba-YOLO-World: Marrying YOLO-World with Mamba for Open-Vocabulary Detection
Wang, H., He, Q., Peng, J., Yang, H., Chi, M., & Wang, Y. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10888334 · Cited by 21
“During the pre-training stage, we adopt the detection and grounding datasets including Objects365 (V1) [36], GQA [37], and Flickr30k [38].”
A Survey of Zero-Shot Object Detection
Cao, W., Yao, X., Xu, Z., Ye, L., Pan, Y., & Ming, Z. (2025) · Big Data Mining and Analytics · doi:10.26599/bdma.2024.9020098 · Cited by 18
A systematic review: object detection
Guleria, A., Varshney, K., Garima, G., & Jindal, S. (2025) · AI & Society · doi:10.1007/s00146-025-02372-0 · Cited by 16
“Some papers identified emerging datasets like Objects365 (Shao et al. 2019) and custom datasets tailored to niche applications (e.g., medical imaging).”
Few-shot object detection with dense-global feature interaction and dual-contrastive learning
Huang, L. F., Dai, S., & He, Z. (2022) · Applied Intelligence · doi:10.1007/s10489-022-04243-3 · Cited by 15
“Because the images of Object365 are more challenging, all methods cannot achieve a very high nAP 50 .”
MethodInfluentialAn Effective Motorcycle Helmet Object Detection Framework for Intelligent Traffic Safety
Cui, S., Zhang, T., Sun, H., Zhou, X., Yu, W., Zhen, A., Wu, Q., & He, Z. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00578 · Cited by 15
“Moreover, we load the pre-trained model parameters on Objects365 [16].”
MethodSelf-citationYang, S. W., Park, H. J., Kim, J. S., Choi, W., Park, J., & Han, S. W. (2023) · Waste Management · doi:10.1016/j.wasman.2023.04.044 · Cited by 14
Detection Hub: Unifying Object Detection Datasets via Query Adaptation on Language Embedding
Meng, L., Dai, X., Chen, Y., Zhang, P., Chen, D., Liu, M., Wang, J., Wu, Z., Yuan, L., & Jiang, Y. (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/cvpr52729.2023.01097 · Cited by 14
“We mainly evaluate our model on three popular object detection benchmarks: COCO [21], Object365 [35] and Visual-Genome [14] following the common practice.”
MethodA New Deep Learning-based Dynamic Paradigm Towards Open-World Plant Disease Detection
Dong, J., Fuentes, A., Yoon, S., Kim, H., Jeong, Y., & Park, D. S. (2023) · Frontiers in Plant Science · doi:10.3389/fpls.2023.1243822 · Cited by 19
“, 2014), Ours (c) used Object-365-v2 (Shao et al., 2019), and Ours (d) used the LVIS (Gupta et al.”
BackgroundMethodGradient-based Sampling for Class Imbalanced Semi-supervised Object Detection
Li, J., Lin, X., Zhang, W., Tan, X., Li, Y., Han, J., Ding, E., Wang, J., & Li, G. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01502 · Cited by 14
“Object365 [30] is a more challenging dataset with 175k images and 365 classes.”
BackgroundSelf-citationGraCo: Granularity-Controllable Interactive Segmentation
Zhao, Y., Li, K., Cheng, Z., Qiao, P., Zheng, X., Ji, R., Liu, C., Li, Y., & Chen, J. (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/cvpr52733.2024.00336 · Cited by 14
MarineInst: A Foundation Model for Marine Image Analysis with Instance Visual Description
Zheng, Z., Chen, Y., Zeng, H., Vu, T.-A., Hua, B., & Yeung, S.-K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72627-9_14 · Cited by 18
“The generated semantics of Semantic-SAM come from pre-defined object categories in existing training datasets [17,48,68] (mainly in-air objects).”
Huang, Y., Wang, X., Liu, X., Cai, L., Feng, X., & Chen, X. (2025) · Agronomy · doi:10.3390/agronomy15051173 · Cited by 17
Luo, G., Yang, X., Dou, W., Wang, Z., Liu, J., Dai, J., Qiao, Y., & Zhu, X. (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/cvpr52734.2025.02324 · Cited by 23
“Restrictions apply.”
DAP: Detection-Aware Pre-training with Weak Supervision
Zhong, Y., Wang, J., Wang, L., Peng, J., Wang, Y.-X., & Zhang, L. (2021) · 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/cvpr46437.2021.00451 · Cited by 14
“This is different from pre-training on a fully-supervised detection data [32, 21], which requires expensive annotation cost.”
BackgroundFindIt: Generalized Localization with Natural Language Queries
Kuo, W., Bertsch, F., Li, W., Piergiovanni, A., Saffar, M. H., & Angelova, A. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20059-5_29 · Cited by 14
Scaling Up Multi-domain Semantic Segmentation with Sentence Embeddings
Yin, W., Liu, Y., Shen, C., Sun, B., & van den Hengel, A. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02060-4 · Cited by 18
“To obtain a robust segmentation model, we merge 9 datasets for training, including 7 well-annotated datasets, a coarsely-annotated dataset (OpenImages [11]), and a weakly annotated dataset (Objects365 [12]).”
InfluentialLarge Batch Optimization for Object Detection: Training COCO in 12 minutes
Wang, T., Zhu, Y., Zhao, C., Zeng, W., Wang, Y., Wang, J., & Tang, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58589-1_29 · Cited by 14
“Nowadays, datasets like Open Images [12] and Objects365 [20] come in an unprecedented scale.”
BackgroundSlender Object Detection: Diagnoses and ImprovementsPreprint
Wan, Z., Chen, Y., Deng, S., Chen, K., Yao, C., & Luo, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.08529 · Cited by 14
“terms of slenderness. of regular objects. Such a bias against slender objects can be mitigated by extending the validation set of COCO. We incorporate slender objects from another dataset, Objects365 [24], to complement COCO. Objects365 is a dataset aiming at object detection in the wild, containing 38k validation ima…”
On Calibration of Object Detectors: Pitfalls, Evaluation and Baselines
Kuzucu, S., Öksüz, K., Sadeghi, J., & Dokania, P. K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72664-4_11 · Cited by 18
Malagoli, E., & Persio, L. D. (2025) · Mathematics · doi:10.3390/math13060893 · Cited by 19
Scaling Novel Object Detection with Weakly Supervised Detection Transformers
LaBonte, T., Song, Y., Wang, X., Vineet, V., & Joshi, N. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00017 · Cited by 15
“A major factor in the success of these approaches is the availability of datasets with fine-grained bounding box and segmentation annotations [20, 43, 38, 27, 59, 39].”
BackgroundTang, C., Xie, L., Zhang, X., Hu, X., & Tian, Q. (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/cvpr52729.2023.01465 · Cited by 15
“Although the datasets for image classification [14] and object detection [38, 30, 48] often have a larger number of images, we consider semantic/instance segmentation in this paper towards accurate, pixel-level recognition results.”
BackgroundSelf-citationAchievement-based Training Progress Balancing for Multi-Task Learning
Yun, H., & Cho, H. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01553 · Cited by 13
“Then, we provided the results for DeepLabV3+ with shared ASPP (Table C), and individual ASPP with DeepLabV3 (Table D) and DeepLabV3+ (Table E).”
ResultOV-VG: A benchmark for open-vocabulary visual grounding
Wang, C., Feng, W., Li, X., Cheng, G., Lyu, S., Liu, B., Chen, L., & Zhao, Q. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127738 · Cited by 15
Merlin: Empowering Multimodal LLMs with Foresight Minds
Yu, E., Zhao, L., Wei, Y., Yang, J., Wu, D., Kong, L., Wei, H., Wang, T., Ge, Z., Zhang, X., et al. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73235-5_24 · Cited by 36
Self-citationOV-Uni3DETR: Towards Unified Open-Vocabulary 3D Object Detection via Cycle-Modality Propagation
Wang, Z., Li, Y., Liu, T., Zhao, H., & Wang, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72970-6_5 · Cited by 18
Domain Adaptation Through Task Distillation
Zhou, B., Kalra, N., & Krähenbühl, P. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58574-7_40 · Cited by 15
“Recognition tasks are carefully hand-designed to infer a compact, general, and abstract representation of the world [7, 12, 27, 29, 47].”
Pan, J., Liu, Y., He, X., Peng, L., Li, J., Sun, Y., & Huang, X. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00143 · Cited by 15
“Our proposed method builds upon GroundingDINO [21], pre-trained on a variety of datasets including MS-COCO [20], Objects365 [36] [36], GoldG [12], Cap4M [21], OpenImages [17], ODinW-35 [19], and RefCOCO [15].”
Self-citation2 nd Workshop on Maritime Computer Vision (MaCVi) 2024: Challenge Results
Kiefer, B., Žust, L., Kristan, M., Perš, J., Teršek, M., Wiliem, A., Messmer, M., Yang, C.-Y., Huang, H.-W., Jiang, Z., et al. (2024) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw60836.2024.00099 · Cited by 13
“We do not record the FPS but according to the original paper of BoTSORT, the FPS is roughly 6.6.”
MethodInfluentialSelf-citationGriffon: Spelling Out All Object Locations at Any Granularity with Large Language Models
Zhan, Y., Zhu, Y., Chen, Z., Yang, F., Tang, M., & Wang, J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72946-1_23 · Cited by 23
OVGNet: A Unified Visual-Linguistic Framework for Open-Vocabulary Robotic Grasping
Li, M., Zhao, Q., Lyu, S., Wang, C., Ma, Y., Cheng, G., & Yang, C. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802654 · Cited by 15
“We initialize the OVGNet parameters with GroundingDINO, which has been pre-trained on diverse datasets including COCO [35], O365 [36], LIVS [37], V3Det [38], Flickr30k [39], and GRIT-200K.”
Text-Guided Object Detection Accuracy Enhancement Method Based on Improved YOLO-World
Ding, Q., Zhang, E., Liu, Z., Yao, X., & Pan, G. (2024) · Electronics · doi:10.3390/electronics14010133 · Cited by 13
PreDet: Large-scale weakly supervised pre-training for detection
Ramanathan, V., Wang, R., & Mahajan, D. K. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00286 · Cited by 12
“On the other hand, Object365 [45] showed that pre-training a detection model on a larger detection dataset leads to performance improvements for detection.”
BackgroundNarrowing the Gap: Improved Detector Training With Noisy Location Annotations
Wang, S., Gao, J. G., Li, B., & Hu, W. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3211468 · Cited by 13
“including PASCAL VOC [1], COCO [2], and Objects365 [3], contribute to rapid advances in modern CNN-based object”
BackgroundFocused Decoding Enables 3D Anatomical Detection by Transformers
Wittmann, B., Navarro, F., Shit, S., & Menze, B. H. (2023) · The Journal of Machine Learning for Biomedical Imaging · doi:10.59275/j.melba.2023-35e6 · Cited by 12
“Over time, object detection has firmed its position as a fundamental task in computer vision (Lin et al., 2014; Shao et al., 2019; Geiger et al., 2012).”
BackgroundVideo OWL-ViT: Temporally-consistent open-world localization in video
Heigold, G., Keysers, D., Minderer, M., Lučić, M., Gritsenko, A. A., Yu, F., Bewley, A., & Kipf, T. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01269 · Cited by 12
“100k images) and Objects365 [29] (approx.”
MethodObject detection survey for industrial applications with focus on quality control
Kühlechner, R. (2025) · Production Engineering · doi:10.1007/s11740-025-01369-4 · Cited by 17
Benchmarking pig detection and tracking under diverse and challenging conditions
Henrich, J., Post, C., Zilke, M., Shiroya, P., Chanut, E., Yamchi, A. M., Yahyapour, R., Kneib, T., & Traulsen, I. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.111264 · Cited by 13
Ma, B., Jia, T., Wang, H., & Chen, D. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3609135 · Cited by 13
“For natural images, there exist many datasets containing millions of images, such as COCO [46], Open Images [47], and Objects365 [48], which provide robust data support for the complex downstream tasks.”
Visual Object Detection with DETR to Support Video-Diagnosis Using Conference Tools
Biró, A., Jánosi-Rancz, K. T., Szilágyi, L. S., Cuesta‐Vargas, A., Martín‐Martín, J., & Szilágyi, S. M. (2022) · Applied Sciences · doi:10.3390/app12125977 · Cited by 11
Pest Detection and Recognition: An approach using Deep Learning Techniques
Kundur, N. C., & Mallikarjuna, P. B. (2022) · International Conference on Cognitive Computing and Information Processing · doi:10.1109/ccip57447.2022.10058692 · Cited by 11
Morawski, I., He, K., Dangi, S., & Hsu, W. H. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00604 · Cited by 11
Huang, R., Cai, K., Han, J., Liang, X., Pei, R., Lu, G., Xu, S., Zhang, W., & Xu, H. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73116-7_9 · Cited by 11
RelationLMM: Large Multimodal Model as Open and Versatile Visual Relationship Generalist
Xie, C., Liang, S., Li, J., Zhang, Z., Zhu, F., Zhao, R., & Wei, Y. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3531452 · Cited by 13
“Object detection datasets: We use the training set of COCO [88] with 82 k images, and a subset of Object365 [89] with 300 k images.”
Self-citationDiagnosing Human-Object Interaction Detectors
Zhu, F., Xie, Y., Xie, W., & Jiang, H. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02369-8 · Cited by 11
A Comprehensive Evaluation of Monocular Depth Estimation Methods in Low-Altitude Forest Environment
Jia, J., Kang, J., Chen, L., Gao, X., Zhang, B., & Yang, G. (2025) · Remote Sensing · doi:10.3390/rs17040717 · Cited by 12
CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection
Chen, Q., Jin, W., Ge, J., Liu, M., Yan, Y., Jiang, J., Yu, L., Guo, X., Li, S., & Chen, J. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i2.32212 · Cited by 11
A Method to Automatic Create Dataset for Training Object Detection Neural Networks
Zhou, S. S., Yang, Z., Zhu, M., Li, H. L., Serikawa, S., Mizumachi, M., & Zhang, L. (2022) · IEEE Access · doi:10.1109/access.2022.3195490 · Cited by 11
Learning to Name Classes for Vision and Language Models
Parisot, S., Yang, Y., & McDonagh, 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/cvpr52729.2023.02248 · Cited by 12
“For this purpose, we build a reference vocabulary of 3031 object names, using the following dataset class names (filtered for duplicates): LVIS (1203 classes), CODA (43 classes), Objects365 [40] (365 classes), ImageNet (1000 classes), and the THINGS [16] database (1854 concepts).”
BackgroundMethodOmniLabel: A Challenging Benchmark for Language-Based Object Detection
Schulter, S., G, V. K. B., Suh, Y., Dafnis, K. M., Zhang, Z., Zhao, S., & Metaxas, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01098 · Cited by 13
“Existing datasets: We start with the validation/test sets of COCO [30], Objects-365 [40], and OpenImages-V5 [21], which not only saves annotation cost for obtaining bounding boxes, but also helps collecting diverse object descriptions.”
BackgroundMethodUSB: Universal-Scale Object Detection BenchmarkPreprint
Shinya, Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.14027 · Cited by 16
“number of categories has been further expanded by recent benchmarks, such as Open Images [32], Objects365 [51], and LVIS [24].”
Background3D Reconstruction of Buildings Based on 3D Gaussian Splatting
Cai, Z., Yang, J., Wang, T., Huang, H., & Guo, Y. (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-w10-2024-37-2024 · Cited by 15
Hyperbolic Learning with Synthetic Captions for Open-World Detection
Kong, F., Chen, Y., Cai, J., & Modolo, D. (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/cvpr52733.2024.01586 · Cited by 11
“For model pre-training, we used two datasets: Object 365 (O365) [45] and GoldG [20], to ensure a fair comparison with other methods on the same evaluation benchmark datasets.”
Method‘Eyes of a Hawk and Ears of a Fox’: Part Prototype Network for Generalized Zero-Shot Learning
Feinglass, J., Thiagarajan, J. J., Anirudh, R., Jayram, T. S., & Yang, Y. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00775 · Cited by 10
“The VINVL [43] detector is extensively pre-trained across Open Images [16], Objects375 [32], COCO [17], Visual Genome [15] with standard object annotations before being fine-tuned on Visual Genome with both an object detection and attribute classification loss has demonstrated State-of-the-Art…”
MethodRoad Object Detection Robust to Distorted Objects at the Edge Regions of Images
Shin, W., Choi, D., Park, H., & Kim, J. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00720 · Cited by 10
“Co-DINO (Swin-L) was pretrained with Objects365 [15] and COCO [10] datasets.”
MethodExploring Multi-Modal Contextual Knowledge for Open-Vocabulary Object Detection
Xu, Y., Zhang, M., Yang, X., & Xu, C. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3485518 · Cited by 17
“In addition, we use Objects365 [71] and OpenIm-ages [72] datasets to conduct cross-dataset evaluation, in order to verify the generalization of our models.”
MethodA Simple Latent Diffusion Approach for Panoptic Segmentation and Mask Inpainting
Gansbeke, W. V., & Brabandere, B. D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72633-0_5 · Cited by 10
Wang, G., Xie, J., Zhang, T., Sun, Y., Chen, H., Zhuang, Y., & Li, J. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3564332 · Cited by 15
“The current research in the object detection predominantly focuses on closed-set detection, where detectors are trained on datasets with predefined categories, such as the COCO [10] and Objects365 [11] datasets.”
Self-citationDomain generalization plant leaf disease recognition: Toward from laboratory to field
Zhan, K., Peng, Y., Liao, M., & Wang, Y. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.111168 · Cited by 13
Learning with Free Object Segments for Long-Tailed Instance Segmentation
Zhang, C., Pan, T.-Y., Chen, T., Zhong, J., Fu, W., & Chao, W. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20080-9_38 · Cited by 12
“Yet, when it comes to rare, less commonly seen objects, there is a drastic performance drop due to insuffcient training examples [6, 18, 28].”
BackgroundAdaptive Enhanced Detection Network for Low Illumination Object Detection
Wei, H., Yu, B., Wang, W., & Zhang, C. (2023) · Mathematics · doi:10.3390/math11102404 · Cited by 10
“The main reason for this limitation is that most of the simulation scenes used in algorithm design are standard scenes [7–10], which makes them unable to adapt to changes and interference caused by low illumination scenes in target texture, structure, and color characteristics [11,12].”
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
