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 401–500 of 677
Large Occluded Human Image Completion via Image-Prior Cooperating
Zhao, H., Zeng, Y., Lu, H., & Wang, L. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i7.28578 · Cited by 6
“As a result, we opted to use the AHP dataset (Zhou et al. 2021), which consists of 56,599 images collected from several large-scale instance segmentation and detection datasets, including COCO (Lin et al. 2014), VOC (Evering-ham et al. 2010), LIP (Gong et al. 2017), Objects365 (Shao et al. 2019) and OpenImages (Kuznet…”
BackgroundQiu, K., Poon, P., Zhao, S., Towey, D., & Yu, L. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112027 · Cited by 4
Learning to Segment Referred Objects from Narrated Egocentric Videos
Shen, Y., Wang, H., Yang, X., Feiszli, M., Elhamifar, E., Torresani, L., & Mavroudi, E. (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/cvpr52733.2024.01375 · Cited by 4
“…backbone) outperforms all compared baselines, including approaches that have been trained with strong spatial supervision ( e.g. , GroundedSAM is based on Ground-ing DINO [27], an open-set object detector that has been pre-trained on Objects365 [40], a large dataset with 10 M bbox annotations).”
Attention Guidance through Video Script: A Case Study of Object Focusing on 360º VR Video Tours
Silva, P. V., Vitória, A. R. S., Silva, D. F. C., & Filho, A. R. G. (2024) · Symposium on Virtual and Augmented Reality (SVR) · doi:10.1145/3691573.3691618 · Cited by 4
“Inspired by GLIP [10], the object detection task is reframed into a phrase grounding task by integrating category names into text prompts sourced from datasets like COCO [11], O365 [20], and OpenImage (OI) [20], with category names dynamically sampled during training for text input variation.”
A Fair Ranking and New Model for Panoptic Scene Graph Generation
Lorenz, J., Pest, A., Kienzle, D., Ludwig, K., & Lienhart, R. W. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73030-6_9 · Cited by 4
Stavrothanasopoulos, K., Gkountakos, K., Ioannidis, K., Tsikrika, T., Vrochidis, S., & Kompatsiaris, I. Y. (2025) · Sensors · doi:10.3390/s25041016 · Cited by 3
VisionKG: Unleashing the Power of Visual Datasets via Knowledge Graph
Yuan, J., Le-Tuan, A., Nguyen-Duc, M., Tran, T.-K., Hauswirth, M., & Le-Phuoc, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-60635-9_5 · Cited by 6
Clustering swap prediction for image-text pre-training
Fayou, S., Ngo, H. C., Sek, Y. W., & Meng, Z. (2024) · Scientific Reports · doi:10.1038/s41598-024-60832-x · Cited by 4
Language-aware multiple datasets detection pretraining for DETRs
Hao, J., & Chen, S. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106506 · Cited by 4
Tiny Models are the Computational Saver for Large Models
Wang, Q., Cardiff, B., Frappé, A., Larras, B., & John, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72992-8_10 · Cited by 4
Ke, J., Wang, D., Chen, J.-C., Jhuo, I., Lin, C., & Lin, Y. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521844 · Cited by 4
“…matching loss, we first compute the similarity scores between the final aggregated features of each node of two graphs (i.e., G a ( T ) and G c ( T ) ) and the textual features of the whole expression (i.e., q ) as follows: where ∗ ∈ { a, c } and W v ∗ and W q are trainable matrices of parameters.”
Islam, N. U., Ma, D., Pang, J., Velan, S. S., Gotway, M. B., & Liang, J. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00359 · Cited by 4
“It then combines various ViM modules (adapters) trained on datasets such as COCO [18], Ob-jects365 [29], and others, to finally finetune on the target downstream dataset.”
LwST-FSD: Efficient forest smoke detection via lightweight spatio-temporal modeling
Wang, C., Song, H., Wang, Z., Shan, Z., & Zhang, Q. (2025) · Information Sciences · doi:10.1016/j.ins.2025.122481 · Cited by 4
CerberusDet: Unified Multi-dataset Object Detection
Tolstykh, I., Chernyshov, M., & Kuprashevich, M. (2025) · Communications in computer and information science · doi:10.1007/978-3-032-01169-5_6 · Cited by 4
Multilingual Communication System with Deaf Individuals Utilizing Natural and Visual Languages
Huynh, T.-L., Nguyen-Ngoc, K.-N., Chu, C.-B., Tran, M., & Le, T.-N. (2022) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf55975.2022.10013851 · Cited by 4
“trained of object detector on four datasets: COCO [33], Visual Genome, Open Images [34], Object365 [35], and applied Parallel Cross-Attention [26] for image captioning.”
MethodA Multi-object Detection Sampling Algorithm For Large Scenes
Jin, L., Li, X., Fan, B., Guo, Z., Li, R., Wang, L., Wang, Y., Zhao, Y., & Li, R. (2022) · International Symposium on Parallel Architectures, Algorithms and Programming · doi:10.1109/paap56126.2022.10010614 · Cited by 4
“Despite above-mentioned methods can successfully find objects in common collected public datasets scenes including Pascal VOC [9], COCO [10], object365 [11], etc.”
MethodData Augmentation Ensemble Module based on Natural Guidance for X-ray Prohibited Items Detection
He, J., Zhong, Y., Sun, B., Zhang, Y., & Liu, Y. (2023) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn54540.2023.10191378 · Cited by 4
ViM: Vision Middleware for Unified Downstream Transferring
Feng, Y., Gong, B., Jiang, J., Lv, Y., Shen, Y., Zhao, D., & Zhou, J. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01074 · Cited by 4
Detection of Defective Products in Injection Molding Process using YOLO-NAS
Asadi, M., Hashemi, S. S., & Sadeghi, M. T. (2023) · International Conference on Control, Instrumentation, and Automation · doi:10.1109/iccia61416.2023.10506371 · Cited by 4
“The model training commences with pre-training on the Objects365 dataset [17], comprising two million images across 365 categories, followed by the generation of pseudo-labels using the COCO dataset [18].”
MethodSC- Tune: Unleashing Self-Consistent Referential Comprehension in Large Vision Language Models
Yue, T., Cheng, J. Z., Guo, L., Dai, X., Zhao, Z., He, X., Xiong, G., Lv, Y., & Liu, 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/cvpr52733.2024.01242 · Cited by 10
“As illustrated in Figure 1, we randomly select 4k bounding boxes from both the in-domain (seen during training) dataset RefCOCO [55] and the out-of-domain (unseen during training) datasets, i.e ., OpenImages [21] and Object365 [40], respectively.”
BackgroundMethodActive learning with sampling by joint global-local uncertainty for salient object detection
Li, L., Fu, H., & Xu, X. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06395-8 · Cited by 4
GLSD: The Global Large-Scale Ship Database and Baseline EvaluationsPreprint
Shao, Z., Wang, J., Deng, L., Huang, X., Lü, T., Luo, F., Zhang, R., Lv, X., Dang, C., Ding, Q., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.02773 · Cited by 8
“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.”
1st Place Solution for the UVO Challenge on Image-based Open-World Segmentation 2021Preprint
Du, Y., Guo, W. J., Xiao, Y., & Lepetit, V. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.10239 · Cited by 6
(2023) · doi:10.18653/v1/2023.emnlp-main.824 · Cited by 4
Noise-Aware Evaluation of Object Detectors
Murrugarra-Llerena, J., & Jung, C. R. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00903 · Cited by 4
“For example, Figure 1 shows the bounding boxes produced by two different human annotators, which were requested to draw bounding boxes from eight classes (keyboard, mouse, monitor, laptop, tablet, cellphone, microphone, vehicle) of an image from the Objects365 dataset [24].”
UAV-OVD: Open-Vocabulary Object Detection in UAV Imagery via Multi-Level Text-Guided Decoding
Tao, L., Wei, G., Wang, Z., Qi, Z., Li, Y., & Zhang, H. (2025) · Drones · doi:10.3390/drones9070495 · Cited by 4
Open vocabulary detection for concealed object detection in AMMW image
Jiang, C., Li, C., & Zhao, X. (2025) · Scientific Reports · doi:10.1038/s41598-025-13935-y · Cited by 4
“Traditional object detectors can only be trained on a predefined set of fixed categories, such as the Objects365 41 dataset and COCO dataset 42 .”
He, Y., Su, H., Yu, H., Yang, C., Wei, S., Wang, C., & Liu, S. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11127317 · Cited by 5
“Following [17], the datasets employed for pre-training comprise: 1) Objects365v1 [30] with 609k images and 9,621k annotations; 2) GQA [31] with 621k images and 3,681k annotations; 3) Flickr30k [32] with 149k images and 641k annotations.”
Oyama, A., Hasegawa, S., Taniguchi, A., Hagiwara, Y., & Taniguchi, T. (2025) · IEEE RO-MAN · doi:10.1109/ro-man63969.2025.11217563 · Cited by 3
“These objects were 567 Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
Quality over quantity: a data-centric survey of annotation errors in object detection datasets
Hussain, A., Ullah, K., Afaq, M., Munsif, M., Hussain, A., & Baik, S. W. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11502-z · Cited by 6
“Object365 is one of the largest datasets featuring 365 different categories, primarily col - lected from Flickr (Shao et al. 2019).”
InfluentialFlyPose: Towards Robust Human Pose Estimation From Aerial Views
Farooq, H., Brenner, M., & Stütz, P. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00831 · Cited by 3
“As a first step, RT-DETRv2-S with pretrained weights from COCO [23] and Objects365 [40] is finetuned for 60 epochs on the VisDrone2019-DET dataset [9] after removing all classes other than “pedestrian” and “person”, which are combined into a single person class.”
Recognizing sensory gestures in historical artworks
Zinnen, M., Hussian, A., Maier, A., & Christlein, V. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-20502-6 · Cited by 3
“Computer Vision and the Humanities With large-scale annotated datasets such as Ima-geNet [12], COCO [13], OpenImages [14], or Objects365 [15], computer vision has significantly advanced in tasks like image classification, object detection or HPE, showing impressive results with natural images.”
STAY Diffusion: Styled Layout Diffusion Model for Diverse Layout-to-Image Generation
Wang, R., Hou, X., Schmedding, S., & Huber, M. F. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00379 · Cited by 3
Towards Real-Time Open-Vocabulary Video Instance Segmentation
Yan, B., Sundermeyer, M., Tan, D. J., Lu, H., & Tombari, F. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00188 · Cited by 3
“In the first stage, our model is pre-trained on two large-scale object detection datasets, Objects365 [35] and OpenImages [20], which contain 1.8M and 1.7M images respectively.”
ViruSeg: Harnessing the Power of Large Language-Image Model for Enhanced Virus Image Segmentation
Wang, S., Fu, X., Du, Z., Liu, X., Mai, Q., Zhuo, L., Xie, B., & Zou, Q. (2025) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-025-00711-9 · Cited by 3
“3 An example of applying cutout augmentation on the TEM images initialized on the COCO [15] and obj365 [16] datasets, were used to fine-tune the downstream virus segmentation task with CMR.”
Fisheye object detection based on standard image datasets with 24-points regression strategy
Xu, X., Gao, Y., Liang, H. Z., Yang, Y. M., & Fu, M. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981891 · Cited by 3
“For example, the number of images in WoodScape [8], a classical fisheye dataset, is less numerous than that of some standard image datasets, such as PASCAL VOC [9], Microsoft COCO [10], Open Images [11] and Objects365 [12].”
BackgroundAttending to Transforms: A Survey on Transformer-based Image Captioning
Ambilduke, K., Jayakumar, T., Farooqui, L., Padole, H. P., & Singh, A. (2023) · International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS) · doi:10.1109/pcems58491.2023.10136098 · Cited by 3
“As the performance of the model directly depends on how well the object detector processes the image, VinVL [29] proposed to use a more refined object detector and pre-trained it on a number of datasets apart from Visual Genome, such as, COCO, OpenImages, Objects365, etc. [7], [35], [36].”
MethodLow-Confidence Samples Mining for Semi-supervised Object Detection
Liu, G., Zhang, F., Pan, T., Yong, J., & Wang, B. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/132 · Cited by 4
Exploring Transformers for Open-world Instance Segmentation
Wu, J., Jiang, Y., Yan, B., Lu, H., Yuan, Z., & Luo, P. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00608 · Cited by 3
“• Extensive experiments demonstrate that our models achieve state-of-the-art performance on several benchmarks including COCO [31], LVIS [16], UVO [47] and Objects365 [39], as shown in Figure 2.”
BackgroundMethodEfficient Transferability Assessment for Selection of Pre-trained Detectors
Zhao, W., Li, A., Li, Z., & Dou, Q. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00170 · Cited by 4
“The object detectors are always trained under supervised [4, 31, 38, 48, 64] and self-supervised [3,11,51,56] schemes on the large-scale datasets [12, 18, 28, 32, 40].”
BackgroundSelf-citationLabeling Indoor Scenes with Fusion of Out-of-the-Box Perception Models
Li, Y., Rajabi, N., Shrestha, S., Reza, M. A., & Košecká, J. (2024) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw60836.2024.00066 · Cited by 3
“The stages are explained in detail below.”
Open-vocabulary object detection via debiased curriculum self-training
Zhang, H., Guan, D., Ke, X., Saddik, A. E., & Lu, S. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124762 · Cited by 3
Towards Unified Interactive Visual Grounding in The Wild
Xu, J., Zhang, H., Si, Q., Li, Y., Lan, X., & Kong, T. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10611354 · Cited by 4
A Coarse-to-fine Two-stage Helmet Detection Method for Motorcyclists
Zhang, H., Cui, Z., & Su, F. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00700 · Cited by 3
“For the Fine-grained detector, we use Co-DETR with Swin-L backbone, which loads the pre-trained model parameters on Objects365 [22].”
MethodPlain-Det: A Plain Multi-dataset Object Detector
Shi, C., Zhu, Y., & Yang, S. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72652-1_13 · Cited by 10
Weak-to-Strong Compositional Learning from Generative Models for Language-Based Object Detection
Park, K., Saito, K., & Kim, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73337-6_1 · Cited by 3
ProMerge: Prompt and Merge for Unsupervised Instance Segmentation
Li, D., & Shin, G. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-72630-9_1 · Cited by 6
Oyama, A., Hasegawa, S., Hagiwara, Y., Taniguchi, A., & Taniguchi, T. (2024) · IEEE International Conference on Robotic Computing (IRC) · doi:10.1109/irc63610.2024.00007 · Cited by 3
“The location of each object was obtained in advance using YOLOv5 [35] trained on Object365 [36].”
Wang, M., Xu, X. D., & Liu, H. (2025) · Sensors · doi:10.3390/s25061798 · Cited by 6
“Relying on large-scale annotated datasets [12–15] to supervise the training of model hyperparameters, the reasoning ability of object detectors is improving without a break.”
Yang, J., Zeng, A., Ren, T., Liu, S., Li, F., Zhang, R., & Zhang, L. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3555527 · Cited by 5
Investigation of Multi-task Learning for Object Detection
Zhang, Y., Li, D., & Xiang, J. (2022) · Lecture notes in electrical engineering · doi:10.1007/978-981-16-6963-7_27 · Cited by 3
Entri: Ensemble Learning with Tri-Level Representations for Explainable Scene RecognitionPreprint
Aminimehr, A., Molaei, A., & Cambria, E. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4482110 · Cited by 3
“5 dataset [71], Yolov5m [72] pretrained on the Objects365 dataset [73], YOLOv5l pre-trained on the Open Images V6 dataset [74], and Yolov5x6 pretrained on the COCO dataset [75] as the object detectors, and has two fully-connected networks with one hidden layer with sizes of 4096 and 8192,…”
MethodBSMH: Cross‐dataset object detection based on box‐separated multiple‐head
Lu, F., Xu, Y. C., Qi, Y., Xie, D. S., & Li, Y. L. (2024) · IET Image Processing · doi:10.1049/ipr2.13152 · Cited by 2
OSR-ViT: A Simple and Modular Framework for Open-Set Object Detection and Discovery
Inkawhich, M. J., Inkawhich, N., Yang, H., Zhang, J., Linderman, R., & Chen, Y. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10826036 · Cited by 2
“The second is to train on the COCO training set and test on 40,000 images from the 365-class Objects365 [55] dataset.”
Generalized Concordant Vision Transformer With Masked Image Tokens for Object Detection
Quan, Y., Zhang, D., & Tang, J. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3570504 · Cited by 2
“We also apply it to a large-scale object detection benchmark, namely Objects365 [64].”
GDVT: geometrically aware dual transformer encoding visual and textual features for image captioning
Chang, Y., Ma, H.-S., Li, S.-C., Jaysawal, B. P., & Huang, J.-W. (2025) · International Journal of Data Science and Analytics · doi:10.1007/s41060-025-00836-6 · Cited by 2
Inv-Adapter: ID Customization Generation via Image Inversion and Lightweight Parameter Adapter
Xing, P., Wang, N., Ouyang, J., & Li, Z. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3590321 · Cited by 4
“( 2)Fine-tuneingInv-Adapter:Subjectedtominimal fine-tuningusingthe3000imagesfromObject365dataset[42] (non-paireddata),enablingrobustpersonalizationofdiverse objects(e.g.,animals,accessories,naturalitems).”
Self-citationTest-Time Vocabulary Adaptation for Language-Driven Object Detection
Liu, M., Hayes, T. L., Mancini, M., Ricci, E., Volpi, R., & Csurka, G. (2025) · Proceedings - International Conference on Image Processing · doi:10.1109/icip55913.2025.11084618 · Cited by 2
Zhang, X., Wulamu, A., Guo, X., & Tohti, T. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-04607-5 · Cited by 4
“Most object detection algorithms are designed for standard datasets like COCO [31], OBJ365 [32], and ImageNet [33], but they’re not suitable for small object detection.”
Self-citationObject Detection Data Synthesis via Box-to-Image Generation Based on Diffusion Models
Zhu, J., Ma, H., Chen, J., & Yuan, J. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3609962 · Cited by 2
MRpilot: A Mixed-Reality System for Responsive Navigation of General Procedural Tasks
Yang, H., Zhou, J., Xu, P., Fu, H., & Huang, H. F. (2025) · Proceedings - International Symposium on Mixed and Augmented Reality, ISMAR · doi:10.1109/ismar67309.2025.00031 · Cited by 2
D-Know: Disentangled Domain Knowledge-Aided Learning for Open-Domain Continual Object Detection
He, B., Yan, C., Kou, Y., Wang, Y., Lv, X., Du, H., & Xie, Y. (2025) · Applied Sciences · doi:10.3390/app152312723 · Cited by 2
Multi-object triangulation and 3D footprint tracking for multi-camera systems
Altoé, G. D., Lucas, M. N., Filho, L. C. C., & Vassallo, R. F. (2025) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.5753/sibgrapi.est.2025.38305 · Cited by 2
EinsPT: Efficient Instance-Aware Pre-Training of Vision Foundation Models
Wang, Z., Tian, Y., Xie, L., Wang, Y., & Ye, Q. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3652371 · Cited by 2
“We evaluate EinsPT by pre-training models on the Objects365 dataset [20], which provides large-scale instance-level annotations, and transferring them to downstream tasks including COCO [21] object detection and instance segmentation, ADE20K [22] semantic segmentation, and Cityscapes [23] street…”
InfluentialIDMR: Towards Instance-Driven Precise Visual Correspondence in Multimodal Retrieval
Liu, B., Bao, Y., Lin, S., Wang, X., Tan, X., Wang, Y., Xie, Y., & Lu, C. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00662 · Cited by 8
“This confirms that our approach effectively captures fine-grained visual correspondence between object instances.”
InfluentialToward Adaptive Open-Set Object Detection via Category-Level Collaboration Knowledge Mining
Ji, Y., Ke, J., He, L., Wang, L., & Gao, X. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3683590 · Cited by 2
“To prevent novel-class leakage from ImageNet [68], as noted in [46], the backbone is implemented with weights pre-trained by DINO [69] on the Objects365 dataset [70].”
Towards Scalable Spatial Intelligence Via 2D-To-3D Data Lifting
Miao, X., Duan, H., Qian, Q., Wang, J., Long, Y., Shao, L., Zhao, D., Xu, R., & Zhang, G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00096 · Cited by 6
“By applying our pipeline to richly annotated 2D datasets such as COCO [68] and Objects365-v2 [90], we create COCO-3D and Objects365-v2-3D , marking the first large-scale expansion of spatial data to ∼ 2M distinct scenes spanning over 300 categories across diverse in-the-wild conditions, including…”
Instance Search via Fusing Hierarchical Multi-level Retrieval and Human-object Interaction Detection
Yang, W., Song, Y., Zhao, Z., & Su, F. (2021) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw54120.2021.00263 · Cited by 2
“couch couch(COCO[13]) phone cell phone(COCO), telephone(Objects365[19]) glass cup(COCO), bottle(COCO), wine glass(COCO) bag backpack(COCO), handbag(COCO), suitcase(COCO), briefcase(Objects365) paper book(COCO), newspaper(Manual Labeling) cloth towel(Objects365), clothing(OpenImage[11]), jacket(OpenImage), coat(OpenIma…”
2D Pose-guided Complete Silhouette Estimation of Human Body in Occlusion
Xu, S., Li, X., & Lin, X. (2022) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr56361.2022.9956436 · Cited by 2
“The AHP dataset collects 56,599 human images from several large-scale instance segmentation and detection datasets, including COCO [28], VOC [29](with SBD [30]), LIP [31], Objects365 [32], and OpenImages [33].”
MethodApplication of Object Detection Models for the Detection of Kitchen Furniture - A Comparison
Stecker, B., & Brandt-Pook, H. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-42508-0_9 · Cited by 2
HowToEat: Exploring Human Object Interaction and Eating Action in Eating Scenarios
Wang, Y., Chen, J., & Yanai, . (2023) · International Workshop on Multimedia Assisted Dietary Management (MADiMa) · doi:10.1145/3607828.3617790 · Cited by 2
“There are many common subtasks in object detection, for example, general object detection [8, 23, 32] focuses on identifying various types of objects in the image (e.g., "cars", "person"); hand detection [11, 16] concentrates on detecting and locating the position of the hand, including key points…”
BackgroundUsing Complex Network Communities to Evaluate the Correctness of Object Detection
Zheng, T., Zhao, S., Qiu, K., Poon, P., & Yu, L. (2023) · International Conference on Dependable Systems and Their Applications (DSA) · doi:10.1109/dsa59317.2023.00106 · Cited by 2
“Complex-task ODs involve a large number (e.g., 80) of object classes, as seen in the COCO [4], VOC [5], Object365 [6] datasets.”
BackgroundTask-decoupled interactive embedding network for object detection
Liu, M., Jiao, J., Li, N., & Pang, M. (2024) · Machine Learning · doi:10.1007/s10994-023-06430-w · Cited by 2
“…proposed method has great generalization ability, we verify its effectiveness on three datasets (Pascal VOC (Everingham et al., 2010, 2015), MS COCO (Lin et al., 2014) and Objects365 (Shao et al., 2019)) and two vision tasks (few-shot object detection and open world object detection) individually.”
Remote Sensing Image Detection Method Based on Contrastive Learning
Zhang, Y., Cheng, Q., Zhu, Z., Jiao, J., Qu, S., & Zhang, F. (2023) · International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP) · doi:10.1109/aiiip61647.2023.00028 · Cited by 2
“Among common object detection networks based on deep learning algorithms, CNN-based networks have great advantages in extracting underlying features[6,7,8,9,10].”
BackgroundKwon, Y. H., Park, S., Luan, T., Oh, S., & Heo, J. (2024) · ASCE International Conference on Computing in Civil Engineering (i3CE) · doi:10.1061/9780784485248.062 · Cited by 2
Using Mixture of Experts to accelerate dataset distillation
Xu, Z. P., & Fu, Z. (2024) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2024.104137 · Cited by 2
FELGA: Unsupervised Fragment Embedding for Fine-Grained Cross-Modal Association
Zhuo, Y., & Li, B. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00554 · Cited by 2
“They require large-scale datasets [1, 3, 35, 36] or even expensive manual annotated datasets [21,41] for pre-training.”
BackgroundMultimodal Information Retrieval
Luo, M., Gokhale, T., Varshney, N., Yang, Y., & Baral, C. R. (2024) · Synthesis lectures on computer vision · doi:10.1007/978-3-031-57816-8_3 · Cited by 2
CribNet: Enhancing Infant Safety in Cribs Through Vision-Based Hazard Detection
Zhu, S., Mathew, A., Hatamimajoumerd, E., Wan, M., Taylor, B. J., Venkatesaramani, R., & Ostadabbas, S. (2024) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg59268.2024.10581871 · Cited by 2
“Despite the advancements in object recognition, methods trained on established datasets like MS-COCO [24], Ob-jects365 [35], and open images detection (OID) [23] often overlook hazardous objects pertinent to specific settings.”
MethodAn unmanned aerial system benchmark object detection dataset for deep learning in outfall surveys
Wu, C., Huang, Y., Yang, H., Yao, L., Liu, Y., & Chen, Z. (2024) · International Journal of Digital Earth · doi:10.1080/17538947.2024.2390443 · Cited by 2
“Objects365 (Shao et al. 2020) and Open Images (Krasin et al. 2016) significantly increased the number of object categories and instances when the number of instances exceeded 10 million.”
Masked AutoDecoder is Effective Multi-Task Vision Generalist
Qiu, H., Huang, J., Gao, P., Lu, L., Zhang, X., & Lu, 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/cvpr52733.2024.01342 · Cited by 3
“We adopt a MIM pre-trained ViT from [14] for comparison with Pix2SeqV2 [7] which is pre-trained on Objects365 [43].”
MethodSelf-citationReal-Time Handgun Detection Using Transformers on Nvidia Jetson AGX Xavier
Bustamante, L. A., & Gutiérrez, J. C. (2024) · Latin American Computer Conference (CLEI) · doi:10.1109/clei64178.2024.10700426 · Cited by 2
“The RT-DETR R18A model has pre-trained weights with Object365 [28], which is a large dataset with 2 million images and COCO [29] with 300K images.”
MethodImage Processing and Deep Learning for Classification of Skin Lesion Image
Lalrinawma, D., Kumar, V. D. A., & Lalchhanhima, R. (2024) · International Conference on Smart Electronics and Communication (ICOSEC) · doi:10.1109/icosec61587.2024.10722560 · Cited by 2
SketchQL Demonstration: Zero-Shot Video Moment Querying with Sketches
Wu, R., Chunduri, P., Shah, D., Aravind, A., Payani, A., Chu, X., Arulraj, J. J. P., & Rong, K. (2024) · Proceedings of the VLDB Endowment · doi:10.14778/3685800.3685892 · Cited by 2
“However, SQL-base interfaces require considerable query specification time from users because translating a semantically meaningful event ( e . g ., left turns) into SQL-like rules on top of low-level primitives ( e . g ., location and angle of bounding boxes) can be challenging.”
Progressive Pseudo Labeling for Multi-Dataset Detection Over Unified Label Space
Ye, K., Huang, Z., Xiong, Y., Gao, Y., Xie, J., & Shen, L. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521841 · Cited by 4
“[2], [3], [4], [5], [6], [7], which is built upon a large number of well-annotated public benchmarks with bounding box annotations, e.g. PASCAL VOC [8], MS-COCO [9], WIDER FACE [10], Object365 [11], etc.”
Rethinking the Sparse End-to-End Multiperson Pose Estimation
Xu, X., Zou, Q., & Li, J. (2025) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2024.3525021 · Cited by 3
“Even our result still is inferior to the best performance of ED-Pose ∗ , since it’s additionally pretrained on a large-scale dataset of Objects365 [30] with 5 feature scales.”
Self-citationNaz, S., & Jabeen, F. (2024) · VAWKUM Transactions on Computer Sciences · doi:10.21015/vtcs.v12i2.1911 · Cited by 3
An Attribute-Enriched Dataset and Auto-Annotated Pipeline for Open Detection
Qi, P., Zhang, Y., Li, W., Hu, Y., & Bai, K. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10890077 · Cited by 2
“The Objects365-Attr is an extension of the Objects365 [14] dataset, offering attribute descriptions that span all its categories.”
IAA: Inner-Adaptor Architecture Empowers Frozen Large Language Model with Multimodal Capabilities
Wang, B., Xie, C., Leng, D., & Yin, Y. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i20.35400 · Cited by 5
GLASS: Guided Latent Slot Diffusion for Object-Centric Learning
Singh, K., Schaub-Meyer, S., & Roth, S. (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/cvpr52734.2025.02670 · Cited by 2
“GLASS further establishes a new SotA FID score among OCL methods for conditional image generation.”
CACP: Context-Aware Copy-Paste to Enrich Image Content for Data Augmentation
Guo, Q., Wang, S., Chang, C.-P., & Rambach, J. R. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00513 · Cited by 2
“For instance, in a foreign object detection task[29], foreign objects may span hundreds of categories, making it impractical to rely on public datasets or manual annotation to prepare the dataset.”
Open-world object detection with multi-dataset image–label matching
Fu, Z., Yuan, S., cao, P., Jing, W. P., Wang, H., Zhang, G., Luo, B., & Zhang, H. (2025) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2025.110742 · Cited by 2
Ingrisch, M. A., Schilling, R. M., Chmielewski, I., & Twieg, S. (2025) · Applied Sciences · doi:10.3390/app152212062 · Cited by 2
Neglected Free Lunch – Learning Image Classifiers Using Annotation Byproducts
Han, D., Choe, J., Chun, S.-Y., Chung, J. J. Y., Chang, M., Yun, S., Song, J. Y., & Oh, S. J. (2023) · IEEE International Conference on Computer Vision · doi:10.1109/iccv51070.2023.01848 · Cited by 5
“For example, Objects365 classes are obtained by labelling the 365 classes along with instance bounding boxes (§3.2.1 in [79]).”
InfluentialRealGraph: A Multiview Dataset for 4D Real-world Context Graph Generation
Lin, H., Chen, Z., Zhang, J., Bai, B., Wang, Y., Huang, R., & Fang, L. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00348 · Cited by 2
“The collection of 2D object detection datasets [15, 42, 31, 43, 21, 62, 63, 53] has contributed to remarkable progress of the object detection in 2D images.”
BackgroundInvestigating and Suggesting the Evaluation Dataset for Image Classification Model
Sivamani, S., Chon, S. I., Choi, D. Y., & Park, J. H. (2020) · IEEE Access · doi:10.1109/access.2020.3024575 · Cited by 4
“However, the COCO dataset has been used in many image classification analysis [30]–[36].”
MethodYang, S. W., Park, H. J., Kim, J. S., Choi, W., Park, J., & Han, S. W. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4181525 · Cited by 2
Learning Adaptive Feature for Object Detection in Low-Light ConditionsPreprint
Fu, G., Chu, H., Tu, X., Huang, Y., Liu, L., He, Z., Zhang, C., Zhang, C., & Zhang, X. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4625592 · Cited by 2
Self-citationExploration of Visual Prompt in Grounded Pre-Trained Open-Set Detection
Chen, Q., Jin, W., Li, S., Liu, M., Yu, L., Jiang, J., & Wang, X. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10447589 · Cited by 2
“Specifically, these methods use text prompt obtained from text encoder [13, 14] to replace the classical parametric classification layer and train the detection model with large-scale object detection or language-image data [15–17] to align its image features and text prompt vectors.”
MethodSynthesizing Environment-Specific People in Photographs
Oštrek, M., O’Sullivan, C., Black, M. J., & Thies, J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73013-9_17 · Cited by 5
“The AHP dataset is compiled from large-scale instance segmentation and detection datasets, including COCO [31], VOC [11], SBD [17], and Objects365 [48].”
Method
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
