</> Code with Papers by Syntology

Papers › Multi-region two-stream R-CNN for action detection

Multi-region two-stream R-CNN for action detection

17 Sep 2016European Conference on Computer Vision (ECVV 2016) 2016 9archive 2025-07-28

Xiaojiang Peng, Cordelia Schmid

We propose a multi-region two-stream R-CNN model for action detection in realistic videos. We start from frame-level action detection based on faster R-CNN [1], and make three contributions: (1) we show that a motion region proposal network generates high-quality proposals , which are complementary to those of an appearance region proposal network; (2) we show that stacking optical flow over several frames significantly improves frame-level action detection; and (3) we embed a multi-region scheme in the faster R-CNN model, which adds complementary information on body parts. We then link frame-level detections with the Viterbi algorithm, and temporally localize an action with the maximum subarray method. Experimental results on the UCF-Sports, J-HMDB and UCF101 action detection datasets show that our approach outperforms the state of the art with a significant margin in both frame-mAP and video-mAP

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action DetectionAction RecognitionRegion ProposalSkeleton Based Action Recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Detection J-HMDB MR-TS R-CNN Frame-mAP 0.5 58.5 #10 of 18 Archive leaderboard report
Action Detection J-HMDB MR-TS R-CNN Video-mAP 0.2 74.3 #10 of 18 Archive leaderboard report
Action Detection J-HMDB MR-TS R-CNN Video-mAP 0.5 73.09 #10 of 18 Archive leaderboard report
Action Detection J-HMDB TS R-CNN Frame-mAP 0.5 56.9 #11 of 18 Archive leaderboard report
Action Detection J-HMDB TS R-CNN Video-mAP 0.2 71.1 #11 of 18 Archive leaderboard report
Action Detection J-HMDB TS R-CNN Video-mAP 0.5 70.6 #11 of 18 Archive leaderboard report
Action Detection UCF Sports MR-TS R-CNN Frame-mAP 0.5 84.52 #2 of 7 Archive leaderboard report
Action Detection UCF Sports MR-TS R-CNN Video-mAP 0.2 94.83 #2 of 7 Archive leaderboard report
Action Detection UCF Sports MR-TS R-CNN Video-mAP 0.5 94.67 #2 of 7 Archive leaderboard report
Action Detection UCF Sports TS R-CNN Frame-mAP 0.5 82.30 #3 of 7 Archive leaderboard report
Action Detection UCF Sports TS R-CNN Video-mAP 0.2 94.82 #3 of 7 Archive leaderboard report
Action Detection UCF Sports TS R-CNN Video-mAP 0.5 94.82 #3 of 7 Archive leaderboard report
Action Detection UCF101-24 TS R-CNN Frame-mAP 0.5 39.94 #14 of 19 Archive leaderboard report
Action Detection UCF101-24 MR-TS R-CNN Frame-mAP 0.5 39.63 #15 of 19 Archive leaderboard report
Action Recognition UCF101 MR Two-Sream R-CNN 3-fold Accuracy 91.1 #71 of 91 Archive leaderboard report
Skeleton Based Action Recognition J-HMDB MR Two-Sream R-CNN Accuracy (RGB+pose) 71.1 #7 of 13 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

ConvolutionFaster R-CNNRPNRoIPoolSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections