The DOI matches its Crossref record; 51 of the 53 references it cites were found in a registry or an index.
Flow-Guided Feature Aggregation for Video Object Detection
Open accessAbstract
Extending state-of-the-art object detectors from image to video is challenging. The accuracy of detection suffers from degenerated object appearances in videos, e.g., motion blur, video defocus, rare poses, etc. Existing work attempts to exploit temporal information on box level, but such methods are not trained end-to-end. We present flow-guided feature aggregation, an accurate and end-to-end learning framework for video object detection. It leverages temporal coherence on feature level instead. It improves the per-frame features by aggregation of nearby features along the motion paths, and thus improves the video recognition accuracy. Our method significantly improves upon strong singleframe baselines in ImageNet VID [33], especially for more challenging fast moving objects. Our framework is principled, and on par with the best engineered systems winning the ImageNet VID challenges 2016, without additional bells-and-whistles. The code would be released.
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
“Indeed, overlooking these factors results in a decrease in the object detection performance in videos.”
From A Systematic Review of Drone Based Road Traffic Monitoring System (2022) · cites this paper as [31]
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 · cited 11×
“The ResNet-50, ResNet-101, and the Aligned-Inception-ResNet models are all pre-trained on ImageNet classification.”
BackgroundMethodInfluentialVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 3×
“State-of-the-art methods for general object detection [10, 30, 26, 5] are mainly based on deep CNNs [22, 36, 40, 14].”
BackgroundMethodImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 3×
“State-of-the-art methods for general object detection [10, 30, 26, 5] are mainly based on deep CNNs [22, 36, 40, 14].”
BackgroundMethodFaster 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 · cited 5×
“In [11], a multistage pipeline called Regions with Convolutional Neural Networks (R-CNN) is proposed for training deep CNN to classify region proposals for object detection.”
BackgroundMethodInfluentialFaster 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
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562 · cited 3×
“State-of-the-art methods for general object detection [10, 30, 26, 5] are mainly based on deep CNNs [22, 36, 40, 14].”
BackgroundMethodImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 12×
“Our approach is evaluated on the large-scale ImageNet VID dataset [33].”
MethodInfluential(2014) · Cited by 33,419
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 · cited 3×
“Recent years have witnessed significant progress in object detection [11].”
BackgroundMethod(2016) · Cited by 31,590
Flagged first, then by citations
2 of the 53 references couldn’t be matched to a known paper.
Research works printed but not linked: 2
- yang, 2016, Efficient object detection from videos
- tran, 2016, Deep end2end voxe12voxel prediction, CVPR Workshop
Cited by
Sorted by citations
Simple Baselines for Human Pose Estimation and Tracking
Xiao, B., Wu, H., & Wei, Y. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01231-1_29 · Cited by 2,278
“d occlusion introduced by video frames. As shown in Fig. 2(c), the detector misses the left black person due to fast motion. Temporal information is often leveraged to generate more robust detections [36,35]. We propose to generate boxes for the processing frame from nearby frames using temporal information expressed…”
BackgroundSelf-citationTSM: Temporal Shift Module for Efficient Video Understanding
Lin, J., Gan, C., & Han, S. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00718 · Cited by 2,209
“We also compare to a strong baseline FGFA [60] that uses optical flow to aggregate the temporal information from 21 frames (past 10 frames and future 10 frames) for offline video detection.”
MethodZhou, X., Koltun, V., & Krähenbühl, P. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58548-8_28 · Cited by 1,428
“posals using bounding box regression. In video object detection, Kang et al. [16,17] feed stacked consecutive frames into the network and do object detection for a whole video segment. And Zhu et al. [59] use flow to warp intermediate features from previous frames to accelerate inference. Our method belongs to this cat…”
MethodVideo Enhancement with Task-Oriented Flow
Xue, T., Chen, B., Wu, J., Wei, D., & Freeman, W. T. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-018-01144-2 · Cited by 1,401
“…(Liu et al 2017), light-field interpolation (Wang et al 2017), novel view synthesis (Zhou et al 2016), eye gaze manipulation (Ganin et al 2016), object detection (Zhu et al 2017), denoising (Wen et al 2017), and super-resolution (Caballero et al 2017; Tao et al 2017; Makansi et al 2017).”
A Survey of Deep Learning-Based Object Detection
Jiao, L., Zhang, F., Liu, F., Yang, S., Li, L., Feng, Z., & Qu, R. (2019) · IEEE Access · doi:10.1109/access.2019.2939201 · Cited by 1,454
“Video detectors [266][267][268][269][270][271][272][273][274][275] exploit temporal contexts to meet this challenge.”
Background
Metadata
- Authors
- Zhu, X., Wang, Y., Dai, J., Yuan, L., & Wei, Y.
- Affiliations
- University of Science and Technology of China (CN)
- Microsoft (United States) (US)
- Venue
- 2017 IEEE International Conference on Computer Vision (ICCV)
- Publisher
- IEEE
- Published
- 2017-10
- Pages
- 408-417
- OpenAlex
- W2964286567
- Cited by
- 729
- References
- 53
- Citations by year
- 2026: 48 · 2025: 57 · 2024: 86 · 2023: 91 · 2022: 102 · 2021: 123 · 2020: 96 · 2019: 75 · 2018: 47 · 2017: 4
- FWCI
- 17.47
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE/CVF International Conference on Computer Vision (ICCV) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- video object detection, optical flow, temporal feature aggregation, movement paths, end-to-end learning, motion blur, fast-moving objects
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
