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 101–200 of 677
WorldScribe: Towards Context-Aware Live Visual Descriptions
Chang, R.-C., Liu, Y., & Guo, A. (2024) · ACM Symposium on User Interface Software and Technology (UIST) · doi:10.1145/3654777.3676375 · Cited by 94
“First, WorldScribe monitors changes in the camera’s orientation using the phone’s inertial measurement unit (IMU).”
MixTeacher: Mining Promising Labels with Mixed Scale Teacher for Semi-Supervised Object Detection
Liu, L., Zhang, B., Zhang, J., Zhang, W., Gan, Z., Tian, G., Zhu, W., Wang, Y., & Wang, C. (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/cvpr52729.2023.00712 · Cited by 61
“However, collecting a large amount of high-quality annotations is infeasible as it is labor-intensive and time-consuming, especially for tasks with complicated annotations such as object detection [24, 32] and segmentation [5, 6].”
BackgroundZhu, Z., Li, X., Zhai, J., & Hu, H. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102385 · Cited by 60
Influence Selection for Active Learning
Liu, Z., Ding, H., Zhong, H., Li, W., Dai, J., & He, C. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00914 · Cited by 57
“The annotation for the datasets [20, 35, 6] of object detection is extremely time-consuming.”
The Challenge of Data Annotation in Deep Learning—A Case Study on Whole Plant Corn Silage
Rasmussen, C. B., Kirk, K., & Moeslund, T. B. (2022) · Sensors · doi:10.3390/s22041596 · Cited by 59
“However, the usage of CNNs introduces challenges in creating image datasets, as it is widely known that models require large amounts of annotated data to train [3].”
BackgroundMethodInfluentialTemporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction
Zong, Z., Jiang, D., Song, G., Xue, Z., Su, J., Li, H., & Liu, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00350 · Cited by 68
“The ViT-L we employ is initialized with the Co-DETR [37] pre-trained on the 2D detection benchmark Objects365 [25].”
Open World Entity Segmentation
Qi, L., Kuen, J., Wang, Y., Gu, J., Zhao, H., Torr, P. H. S., Lin, Z., & Jia, J. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3227513 · Cited by 62
“Please refer to the supplementary file for many more cross-dataset qualitative results on ADE20K [90], Cityscapes [12], Places2 [89], and Objects365 [63].”
BackgroundConvex-Hull Feature Adaptation for Oriented and Densely Packed Object Detection
Guo, Z., Zhang, X., Liu, C., Ji, X., Jiao, J., & Ye, Q. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3140248 · Cited by 58
Self-citationThe effectiveness of MAE pre-pretraining for billion-scale pretraining
Singh, M., Duval, Q., Alwala, K. V., Fan, H., Aggarwal, V., Adcock, A., Joulin, A., Dollár, P., Feichtenhofer, C., Girshick, R., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00505 · Cited by 70
“Lastly, methods which use additional detection data, such as Objects365 [67] or FLOD-9M [82], have strong detection performance.”
MethodDeep transfer learning for image classification: a survey
Plested, J., Phiri, M., & Gedeon, T. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11491-z · Cited by 107
“Multi-step fine-tuning pipelines can be particularly effective in the case of small target datasets with limited closely related data [4, 47], as can self-training [35].”
Shin, A., Ishii, M., & Narihira, T. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-021-01547-8 · Cited by 58
“2020), and as such, they first attempt to enhance their visual features by performing large-scale image classification and object detection with Objects365 dataset (Shao et al. 2019).”
BackgroundDetection Transformer with Stable Matching
Liu, S., Ren, T., Chen, J., Zeng, Z., Zhang, H., Li, F., Li, H., Huang, J. G., Su, H., Zhu, J., et al. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00597 · Cited by 75
Going Denser with Open-Vocabulary Part Segmentation
Sun, P., Chen, S., Zhu, C., Xiao, F., Luo, P., Xie, S., & Yan, Z. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01417 · Cited by 68
“Recent advances in open-vocabulary object detection [39,44,49,60,80,87,94,97] have made surprising development in enlarging the number of object categories from a pre-determined set by training datasets [19,28,51,73] to any object in the open world.”
BackgroundT-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy
Jiang, Q., Li, F., Zeng, Z., Ren, T., Liu, S., & Zhang, L. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73414-4_3 · Cited by 70
Learning to Detect Every Thing in an Open World
Saito, K., Hu, P., Darrell, T., & Saenko, K. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20053-3_16 · Cited by 52
“For this purpose, we adopt either COCO [30] or Cityscapes [9] as a training source, with UVO [41], Obj365 [37] and Mappilary Vista [33] as the test datasets.”
MethodTowards All-in-One Pre-Training via Maximizing Multi-Modal Mutual Information
Su, W., Zhu, X., Tao, C., Lu, L., Li, B., Huang, G. M., Qiao, Y., Wang, X., Zhou, J., & Dai, J. (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/cvpr52729.2023.01525 · Cited by 58
Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models
Ma, C., Jiang, Y., Wu, J., Yuan, Z., & Qi, X. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72658-3_24 · Cited by 85
EdaDet: Open-Vocabulary Object Detection Using Early Dense Alignment
Shi, C., & Yang, S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01441 · Cited by 53
“To evaluate the effectiveness of our EdaDet, we conduct experiments on LVIS [17], COCO [29], and Ob-jects365 [42]”
MethodSDDGR: Stable Diffusion-Based Deep Generative Replay for Class Incremental Object Detection
Kim, J., Cho, H., Kim, J., Tiruneh, Y. Y., & Baek, 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/cvpr52733.2024.02718 · Cited by 55
“For incorporating grounding input, we employ pre-trained GLIGEN’s gated self-attention weights that have been trained on various datasets including GoldG [30], O365 [55], SBU [42], and CC3M [56].”
MethodOmni-DETR: Omni-Supervised Object Detection with Transformers
Wang, P., Cai, Z., Yang, H., Swaminathan, G., Vasconcelos, N. M., Schiele, B., & Soatto, S. (2022) · 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/cvpr52688.2022.00915 · Cited by 49
“Most of the successes of recent object detection are attributed to the large-scale well-established object detection datasets [11, 12, 25, 28, 38], which have accurate and complete detection annotation (category and bounding box or segmentation mask) for every object of interest in an image.”
InfluentialRLIPv2: Fast Scaling of Relational Language-Image Pre-training
Yuan, H., Zhang, S., Wang, X., Albanie, S., Pan, Y., Feng, T., Jiang, J., Ni, D., Zhang, Y., & Zhao, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01979 · Cited by 53
“To pre-train RLIPv2, we utilize VG [38], COCO [53] and Objects365 [69].”
BackgroundMethodSynchronizing Object Detection: Applications, Advancements and Existing Challenges
Hosain, M. T., Zaman, A., Abir, M. R., Akter, S., Mursalin, S., & Khan, S. S. (2024) · IEEE Access · doi:10.1109/access.2024.3388889 · Cited by 53
“It serves as a valuable resource for feature learning in object detection and semantic segmentation tasks, outperforming ImageNet models on the COCO benchmark [143].”
MethodGAIA: A Transfer Learning System of Object Detection that Fits Your Needs
Bu, X., Peng, J., Yan, J., Tan, T., & Zhang, Z. (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/cvpr46437.2021.00034 · Cited by 46
“With the advent of larger datasets of object detection such as Objects365 [53], Open Images [34], and Robust Vision [1], there are works [46] focusing on solving”
BackgroundInfluentialCo-mining: Self-Supervised Learning for Sparsely Annotated Object Detection
Wang, T., Yang, T., Cao, J., & Zhang, X. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i4.16385 · Cited by 46
“Most of object detectors are trained in the completely annotated object datasets (Lin et al. 2014; Everingham et al. 2010; Shao et al. 2019).”
BackgroundSelf-citationXu, Y., Sun, Y., Yang, Z., Miao, J., & Yang, Y. (2022) · 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/cvpr52688.2022.01393 · Cited by 56
“Modern object detection methods [6,40, 45] have achieved promising detection accuracy based on some large-scale datasets [17,21,41,52].”
BackgroundÖksüz, K., Çam, B. C., Kalkan, S., & Akbaş, E. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3130188 · Cited by 49
“for evaluating performance on many visual detection tasks and competitions [1], [2], [3], [4], [5], [6], [7].”
BackgroundDetCLIPv3: Towards Versatile Generative Open-Vocabulary Object Detection
Yao, L., Pi, R., Han, J., Liang, X., Xu, H., Zhang, W., Li, Z., & Xu, 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/cvpr52733.2024.02586 · Cited by 54
“DetCLIPv3’s training leverages datasets from multiple sources, including detection [46, 51], grounding [22], and image-text pairs [4, 44, 48, 49] with bounding-box pseudo-labels (as detailed in Sec.”
InfluentialSelf-citationEnhanced Spatial Feature Learning for Weakly Supervised Object Detection
Wu, Z., Wen, J., Xu, Y. D., Yang, J. J., Li, X., & Zhang, D. D. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3178180 · Cited by 44
“However, its training requires large-scale datasets with accurate bounding-box annotations [10]–[12], which are time-consuming and laborious to be obtained.”
BackgroundMini-InternVL: a flexible-transfer pocket multi-modal model with 5% parameters and 90% performance
Gao, Z., Chen, Z., Cui, E., Ren, Y., Wang, W., Zhu, J., Tian, H., He, J., He, J., Zhu, X., et al. (2024) · Visual Intelligence · doi:10.1007/s44267-024-00067-6 · Cited by 84
“Ultimately, we develop a vision encoder, termed InternViT-300M, which is infused with diverse knowledge and is adaptable to various language models.”
Skeleton-Based Mutually Assisted Interacted Object Localization and Human Action Recognition
Xu, L., Lan, C., Zeng, W., & Lu, C. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3175374 · Cited by 42
“In this paper, we employ the Cascade RCNN [72] model trained on Object365 [73] to localize and detect objects.”
BackgroundMethodVary: Scaling up the Vision Vocabulary for Large Vision-Language Model
Wei, H., Kong, L., Chen, J., Zhao, L., Ge, Z., Yang, J., Sun, J., Han, C., & Zhang, X. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73235-5_23 · Cited by 50
Self-citationBoosting Weakly Supervised Object Detection with Progressive Knowledge Transfer
Zhong, Y., Wang, J., Peng, J., & Zhang, L. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58574-7_37 · Cited by 44
“, COCO [18], ImageNet [4], Open Images [15], and Object 365 [23].”
BackgroundX-DETR: A Versatile Architecture for Instance-wise Vision-Language Tasks
Cai, Z., Kwon, G., Ravichandran, A., Bas, E., Tu, Z., Bhotika, R., & Soatto, S. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20059-5_17 · Cited by 48
Sequence-Dropout Block for Reducing Overfitting Problem in Image Classification
Qian, L., Hu, L., Zhao, L., Wang, T., & Jiang, R. (2020) · IEEE Access · doi:10.1109/access.2020.2983774 · Cited by 45
“Thus, a lot of datasets are built such as ImageNet [23], COCO [24], and Objects365 [25].”
BackgroundMulti-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving
Kong, L., Xu, X., Ren, J., Zhang, W., Pan, L., Chen, K., Ooi, W. T., & Liu, Z. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3535625 · Cited by 43
“Over the past few years, image segmentation models have become both more efficient and highly accurate on large-scale benchmarks [2], [91], [92], [93], [94].”
Multi-Label Supervised Contrastive Learning
Zhang, P., & Wu, M. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i15.29619 · Cited by 38
MosaicOS: A Simple and Effective Use of Object-Centric Images for Long-Tailed Object Detection
Zhang, C., Pan, T.-Y., Li, Y., Hu, H., Xuan, D., Changpinyo, S., Gong, B., & Chao, W. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00047 · Cited by 41
“With the rapid developments of neural networks [31, 62, 64, 42] and the efforts in data curation [51, 44, 61, 80, 70], stateof-the-art algorithms [30, 59, 52, 56, 50, 49, 2, 89] have achieved remarkable progress in detecting common objects (e.”
Multiple Instance Detection Networks With Adaptive Instance Refinement
Wu, Z., Wen, J., Xu, Y. D., Yang, J. J., & Zhang, D. D. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3125130 · Cited by 39
“R ECENTLY, fully supervised object detection (FSOD) [1]– [13] has achieved great success due to the development of convolutional neural networks (CNN) and the availability of large-scale datasets with instance-level annotations [14], [15].”
BackgroundHallucination of Multimodal Large Language Models: A Survey
Bai, Z., Wang, P., Xiao, T., He, T., Han, Z., Zhang, Z., & Shou, M. Z. (2024) · arXiv · doi:10.48550/arxiv.2404.18930 · Cited by 38
Zhang, M., Zhang, C., Zhang, Q., Li, Y., Gao, X., & Zhang, J. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3680609 · Cited by 37
“Semantic-SAM is trained using seven datasets, i . e ., SA-1B, COCO panoptic [38], ADE20k panoptic [92], PASCAL part [6], PACO [47], PartIma-geNet [21], and Objects365 [51].”
MethodSDPDet: Learning Scale-Separated Dynamic Proposals for End-to-End Drone-View Detection
Yin, N., Liu, C., Tian, R., & Qian, X. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3371892 · Cited by 36
“1109/TMM.2024.3371892 detection datasets (e.g., MS COCO [4], Pascal VOC [5], Ob-jects365 [6]), object detection has developed at a rapid pace and has a wide range of applications in various fields [7], [8], [9].”
BackgroundInfluentialAligning and Prompting Everything All at Once for Universal Visual Perception
Shen, Y., Fu, C., Chen, P., Zhang, M., Li, K., Sun, X. W., Wu, Y., Lin, S., & Ji, R. (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/cvpr52733.2024.01253 · Cited by 45
“1, we evaluate on the well-established benchmarks, including large-vocabulary and long-tailed dataset, e.g ., LVIS [19], and common object detection datasets, e.g ., Objects365 [40], OpenImages [41] and MSCOCO [39].”
MethodMosaicFusion: Diffusion Models as Data Augmenters for Large Vocabulary Instance Segmentation
Xie, J., Li, W., Li, X., Liu, Z., Ong, Y.-S., & Loy, C. C. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02223-3 · Cited by 48
BigDetection: A Large-scale Benchmark for Improved Object Detector Pre-training
Cai, L., Zhang, Z., Zhu, Y., Zhang, L., Li, M., & Xue, X. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00524 · Cited by 39
“We first give a brief review on three existing large-scale object detection datasets LVIS [24], OpenImages [26] and Objects365 [44].”
BackgroundMethodSelf-citationToğaçar, M., Ergen, B., & Tümen, V. (2022) · Journal of Applied Biomedicine · doi:10.1016/j.bbe.2022.05.005 · Cited by 36
CapDet: Unifying Dense Captioning and Open-World Detection Pretraining
Long, Y., Wen, Y., Han, J., Xu, H., Ren, P., Zhang, W., Zhao, S., & Liang, X. (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/cvpr52729.2023.01462 · Cited by 39
Ramalingam, S. P., & Kumar, V. (2024) · Building and Environment · doi:10.1016/j.buildenv.2024.112174 · Cited by 35
Human De-occlusion: Invisible Perception and Recovery for Humans
Zhou, Q., Wang, S., Wang, Y., Huang, Z., & Wang, X. (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/cvpr46437.2021.00369 · Cited by 36
“It is noticeable that the preserved ratios of OpenImages and Objects365 are significantly smaller than COCO.”
MethodEfficient Image Captioning for Edge Devices
Wang, N., Xie, J., Luo, H., Cheng, Q., Wu, J., Jia, M., & Li, L. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i2.25359 · Cited by 39
Low-Light Image Enhancement Framework for Improved Object Detection in Fisheye Lens Datasets
Tran, D. Q., Aboah, A., Jeon, Y., Shoman, M., Park, M., & Park, S. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00699 · Cited by 34
Advancing building facade solar potential assessment through AIoT, GIS, and meteorology synergy
Dong, K., Yu, Q., Guo, Z., Xu, J. X., Tan, H., Zhang, H., & Yan, J. (2025) · Advances in Applied Energy · doi:10.1016/j.adapen.2025.100212 · Cited by 36
NTIRE 2025 Challenge on Cross-Domain Few-Shot Object Detection: Methods and Results
Fu, Y., Qiu, X., Ren, B., Fu, Y., Timofte, R., Sebe, N., Yang, M., Gool, L. V., Zhang, K., Nong, Q., et al. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00103 · Cited by 34
Self-citationThe 1st Tiny Object Detection Challenge: Methods and Results
Yu, X., Han, Z., Gong, Y., Jan, N., Zhao, J. C., Ye, Q., Chen, J., Feng, Y., Zhang, B., Wang, X., et al. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-68238-5_23 · Cited by 34
“And for tiny[2, 20], it is partitioned into 3 sub-intervals: tiny1[2, 8], tiny2[8, 12], tiny3[12, 20].”
BackgroundMethodA Review of Transformer-Based Approaches for Image Captioning
Ondeng, O., Absaloms, H. O., & Akuon, P. O. (2023) · Applied Sciences · doi:10.3390/app131911103 · Cited by 39
Interactive Story Visualization with Multiple Characters
Gong, Y., Pang, Y., Cun, X., Xia, M., He, Y., Chen, H., Wang, L., Zhang, Y., Wang, X., Shan, Y., et al. (2023) · SIGGRAPH Asia Conference Papers · doi:10.1145/3610548.3618184 · Cited by 77
“For the training of the T2L component, we use the Ob-ject365 dataset [Shao et al. 2019] that contains 365 classes, 2 million images, and 30 million bounding boxes.”
MethodInfluentialVIGC: Visual Instruction Generation and Correction
Wang, B., Wu, F., Han, X. H., Peng, J., Zhong, H., Zhang, P., Dong, X., Li, W., Li, W., Wang, J., et al. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i6.28338 · Cited by 65
Point Cloud Pre-training with Natural 3D Structures
Yamada, R., Kataoka, H., Chiba, N., Domae, Y., & Ogata, T. (2022) · 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/cvpr52688.2022.02060 · Cited by 36
“Therefore, we can easily assume that current point cloud datasets contain a more limited number of data and annotations for training, validation, and test sets as compared to 2D vision datasets [18,33,53,72].”
BackgroundLearning Background Prompts to Discover Implicit Knowledge for Open Vocabulary Object Detection
Li, J., Zhang, J., Li, J., Li, G., Liu, S., Lin, L., & Li, G. (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/cvpr52733.2024.01578 · Cited by 31
“This can enhance the capacity of the detecter to recognize both base and novel categories during inference.”
BackgroundSelf-citationLearning with Noisy Class Labels for Instance Segmentation
Yang, L., Meng, F., Li, H., Wu, Q., & Cheng, Q. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58568-6_3 · Cited by 31
“Mean-while, some existing papers [11,19,24] also mention that inherent ambiguity of classes and limited experience of annotators can result in corrupted object class labels.”
BackgroundUni-Perceiver v2: A Generalist Model for Large-Scale Vision and Vision-Language Tasks
Li, H., Zhu, J., Jiang, X., Zhu, X., Li, H., Yuan, C., Wang, X., Qiao, Y., Wang, X., Wang, W., et al. (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/cvpr52729.2023.00264 · Cited by 41
“Swin-Large is firstly pre-trained on ImageNet-21k and then trained on the detection task with Object365 [32].”
MethodV3Det: Vast Vocabulary Visual Detection Dataset
Wang, J., Zhang, P., Chu, T., Cao, Y., Zhou, Y., Wu, T., Wang, B., He, C., & Lin, D. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01817 · Cited by 53
“1: V3Det is a vast vocabulary visual detection dataset with precisely annotated bounding boxes of 13,204 categories, significantly more diverse than existing visual detection datasets, e.g., COCO [34], Objects365 [45], OpenImages [29], and LVIS [23].”
BackgroundMethodInfluentialSurvey on leveraging pre-trained generative adversarial networks for image editing and restoration
Liu, M., Wei, Y., Wu, X., Zuo, W., & Zhang, L. (2023) · Science China Information Sciences · doi:10.1007/s11432-022-3679-0 · Cited by 38
“Therefore, some other datasets like Objects 365 [66], Places 365 [67], and Open Images [68] may also be used for training GAN models with greater capacity.”
MethodOpen-vocabulary Attribute Detection
Bravo, M. A., Mittal, S., Ging, S., & Brox, T. (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/cvpr52729.2023.00680 · Cited by 36
“[43] Shuai Shao, Zeming Li, Tianyuan Zhang, Chao Peng, Gang Yu, Xiangyu Zhang, Jing Li, and Jian Sun.”
BackgroundImage amodal completion: A survey
Ao, J., Ke, Q., & Ehinger, K. A. (2023) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2023.103661 · Cited by 29
Fine-Grained Visual–Text Prompt-Driven Self-Training for Open-Vocabulary Object Detection
Long, Y., Han, J., Huang, R., Xu, H., Zhu, Y., Xu, C., & Liang, X. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3293484 · Cited by 29
“We evaluate our VTP-OVD on the popular object detection dataset COCO [2] under the well-known zero-shot setting and further validate its generalization capability on the PASCAL VOC [1], Objects365 [16], and LVIS [3] benchmarks.”
MethodMiniVLM: A Smaller and Faster Vision-Language ModelPreprint
Wang, J., Hu, X., Zhang, P., Li, X., Wang, L., Zhang, L., Gao, J., & Liu, Z. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.06946 · Cited by 59
“First, we use the Objects365 dataset to pretrain TEE before fine-tuning it on VG, which improves the CIDEr score by 1.7, indicating that better visual fea- tures contribute to better performance on VL tasks.”
BackgroundMethodInfluentialInstaGen: Enhancing Object Detection by Training on Synthetic Dataset
Feng, C., Zhong, Y., Jie, Z., Xie, W., & Ma, L. (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/cvpr52733.2024.01339 · Cited by 30
Pink: Unveiling the Power of Referential Comprehension for Multi-modal LLMs
Xuan, S., Guo, Q., Yan, S., & Zhang, 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/cvpr52733.2024.01313 · Cited by 45
“The model used to generate referring-expression-bounding-box pairs in Object365 is trained with the aforementioned datasets, excluding Object365 itself.”
MethodInfluentialToward the unification of generative and discriminative visual foundation model: a survey
Liu, X., Zhou, T., Wang, C., Wang, Y., Wang, Y., Cao, Q., Du, W., Yang, Y., He, J., Qiao, Y., et al. (2024) · The Visual Computer · doi:10.1007/s00371-024-03608-8 · Cited by 39
Tokenize Anything via Prompting
Pan, T., Tang, L., Wang, X., & Shan, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72970-6_19 · Cited by 40
PSALM: Pixelwise SegmentAtion with Large Multi-modal Model
Zhang, Z., Ma, Y., Zhang, E., & Bai, X. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72754-2_5 · Cited by 53
Simple Image-Level Classification Improves Open-Vocabulary Object Detection
Fang, R., Pang, G., & Bai, X. P. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i2.27939 · Cited by 27
Ding, H., Seenivasan, L., Killeen, B. D., Cho, S. M., & Unberath, M. (2024) · Artificial Intelligence Surgery · doi:10.20517/ais.2024.16 · Cited by 27
On the Performance of Crowd-Specific Detectors in Multi-Pedestrian Tracking
Stadler, D., & Beyerer, J. (2021) · IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) · doi:10.1109/avss52988.2021.9663836 · Cited by 26
AlignDet: Aligning Pre-training and Fine-tuning in Object Detection
Li, M., Wu, J., Wang, X., Chen, C., Qin, J., Xiao, X., Wang, R., Zheng, M., & Pan, X. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00632 · Cited by 30
“These discrepancies potentially bring limited results, poor generalization, and slow convergence speed [43, 47].”
BackgroundFast-iTPN: Integrally Pre-Trained Transformer Pyramid Network With Token Migration
Tian, Y., Xie, L., Qiu, J., Jiao, J., Wang, Y., Tian, Q. C., & Ye, Q. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3429508 · Cited by 35
“The Fast-iTPN-B model reports 61.8% mAP on COCO and 58.2 mIoU on ADE20 K (pre-trained on Objects365 [9]), surpassing the competitors by large margins.”
MethodInfluentialOmniTracker: Unifying Visual Object Tracking by Tracking-With-Detection
Wang, J., Wu, Z., Chen, D., Luo, C., Dai, X., Yuan, L., & Jiang, Y. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3529926 · Cited by 47
“UNINEXT [62] adopts a more advanced detector DINO [65] than OmniTracker, and first pre-train the model on a large-scale object detection dataset Objects365 [84] for learning universal knowledge about objects.”
Multimodal Autoregressive Pre-training of Large Vision Encoders
Fini, E., Shukor, M., Li, X., Dufter, P., Klein, M., Haldimann, D., Aitharaju, S., da Costa, V. G. T., Béthune, L., Gan, Z., et al. (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/cvpr52734.2025.00901 · Cited by 71
“All models were trained on the following datasets: Objects365v1 [101], Flickr-30k Entities [90, 128], GQA [52], COCO17 [71], and RefCOCO [57, 79, 131].”
Number of Building Stories Estimation from Monocular Satellite Image Using a Modified Mask R-CNN
Ji, C., & Tang, H. (2020) · Remote Sensing · doi:10.3390/rs12223833 · Cited by 25
“In Section 4.3, we will discuss the decision basis of the model, which is maybe the most controversial part of the study.”
BackgroundInfluentialVisual-Language Navigation Pretraining via Prompt-based Environmental Self-exploration
Liang, X., Zhu, F., Lingling, L., Xu, H., & Liang, X. (2022) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2022.acl-long.332 · Cited by 33
Street-view imagery guided street furniture inventory from mobile laser scanning point clouds
Zhou, Y., Xu, H., Peng, M., Li, H., Yang, B., Dong, Z., & Yang, B. (2022) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2022.04.023 · Cited by 25
Object detection in hospital facilities: A comprehensive dataset and performance evaluation
Hu, D., Li, S., & Wang, M. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106223 · Cited by 25
ScaleDet: A Scalable Multi-Dataset Object Detector
Chen, Y., Wang, M., Mittal, A., Xu, Z., Favaro, P., Tighe, J., & Modolo, D. (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/cvpr52729.2023.00704 · Cited by 26
“We show that ScaleDet can boost its performance as we increase the number of training datasets: LVIS [14], COCO [27], Objects365 [34] and OpenImages [24] (Sec 4.”
BackgroundMethodLearning to Detect and Segment for Open Vocabulary Object Detection
Wang, T. (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/cvpr52729.2023.00681 · Cited by 26
“We also evaluate the generalization through inference on two other common object benchmarks, PASCAL VOC [8] and Objects365 [27].”
MethodContrastive Feature Masking Open-Vocabulary Vision Transformer
Kim, D., Angelova, A., & Kuo, W. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01430 · Cited by 25
“To assess CFM-ViT’s ability to generalize in zero-shot transfer detection, we test its performance on Objects365-v1 validation split [52].”
MethodTowards Building Self-Aware Object Detectors via Reliable Uncertainty Quantification and Calibration
Öksüz, K., Joy, T., & Dokania, P. K. (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/cvpr52729.2023.00894 · Cited by 27
Zhao, B., & Song, R. (2024) · Scientific Reports · doi:10.1038/s41598-024-55570-z · Cited by 24
A Robust Deformable Linear Object Perception Pipeline in 3D: From Segmentation to Reconstruction
Zhaole, S., Zhou, H., Li, N., Chen, L., Zhu, J., & Fisher, R. B. (2023) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2023.3337695 · Cited by 28
“Grounding-DINO is pre-trained on large detection datasets, such as O365 [27], which contain ropes as well as other common objects.”
BackgroundUnsupervised Recognition of Unknown Objects for Open-World Object Detection
Fang, R., Pang, G., Miao, W., Bai, X. P., Zheng, J. G., & Ning, X. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3559940 · Cited by 29
“By default, we use ResNet-50 [61] with FPN [62] as the backbone network for feature extraction Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
Vision Transformers are Good Mask Auto-Labelers
Lan, S., Yang, X., Yu, Z., Wu, Z., Álvarez, J. M., & Anandkumar, A. (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/cvpr52729.2023.02274 · Cited by 23
Bridging Precision and Confidence: A Train-Time Loss for Calibrating Object Detection
Munir, M. A., Khan, M. H., Khan, S., & Khan, F. 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/cvpr52729.2023.01104 · Cited by 26
“We opt Object365 [31] validation dataset in case of MS-COCO with similar categories, subset of BDD100k train set for Cityscapes and for Sim10k, its validation split.”
MethodNetTrack: Tracking Highly Dynamic Objects with a Net
Zheng, G., Lin, S., Zuo, H., Fu, C., & Pan, 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/cvpr52733.2024.01811 · Cited by 27
“The default detector is Ground-ingDINO [41] with Swin-B [43] backbone, which was pre-trained on COCO [36], O365 [58], etc .”
MethodZero-Shot Referring Expression Comprehension via Structural Similarity Between Images and Captions
Han, Z., Zhu, F., Lao, Q., & Jiang, H. (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/cvpr52733.2024.01362 · Cited by 30
“Representative datasets contain MS-COCO [37], Object365 [49] (fixed-category), Flickr30K Entities [45] (open-vocabulary), etc.”
GRiT: A Generative Region-to-Text Transformer for Object Understanding
Wu, J., Wang, J., Yang, Z., Gan, Z., Liu, Z., Yuan, J., & Wang, L. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72989-8_12 · Cited by 22
Golroudbari, A. A., & Sabour, M. H. (2023) · doi:10.22541/au.168664884.43899660/v1 · Cited by 22
Robust Data Augmentation and Ensemble Method for Object Detection in Fisheye Camera Images
Duong, V. H., Nguyen, D. Q., Luong, T. V., Vu, H., & Nguyen, T. C. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00695 · Cited by 22
“CO-DETR In our study, we employ the CO-DETR model with the Swin-L backbone architecture, which was pre-trained with the COCO [18] and the Objects365 [33] datasets.”
When Pedestrian Detection Meets Multi-modal Learning: Generalist Model and Benchmark Dataset
Zhang, Y., Zeng, W., Jin, S., Qian, C., Luo, P., & Liu, W. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73195-2_25 · Cited by 26
The Challenge of Variable Effort Crowdsourcing and How Visible Gold Can Help
Hettiachchi, D., Schaekermann, M., McKinney, T., & Lease, M. (2021) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3476073 · Cited by 21
“Secondly, if we imagine training a detection model on this data, it is particularly important to have accurately annotated images with larger object counts [71].”
Rethinking Few-Shot Object Detection on a Multi-Domain Benchmark
Lee, K., Yang, H., Chakraborty, S., Cai, Z., Gurumurthy, S., Ravichandran, A., & Dabeer, O. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20044-1_21 · Cited by 23
Advancing Visual Grounding with Scene Knowledge: Benchmark and Method
Chen, Z., Zhang, R., Song, Y., Wan, X., & Li, G. (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/cvpr52729.2023.01444 · Cited by 26
“Besides, PASCAL VOC 2007 [12], Visual Genome [18], and Object365 [34] are also popular datasets of this type.”
Background
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
