Papers › Representation Flow for Action Recognition

Representation Flow for Action Recognition

2 Oct 2018CVPR 2019 6arXiv:1810.01455archive 2025-07-28

AJ Piergiovanni, Michael S. Ryoo

In this paper, we propose a convolutional layer inspired by optical flow algorithms to learn motion representations. Our representation flow layer is a fully-differentiable layer designed to capture the `flow' of any representation channel within a convolutional neural network for action recognition. Its parameters for iterative flow optimization are learned in an end-to-end fashion together with the other CNN model parameters, maximizing the action recognition performance. Furthermore, we newly introduce the concept of learning `flow of flow' representations by stacking multiple representation flow layers. We conducted extensive experimental evaluations, confirming its advantages over previous recognition models using traditional optical flows in both computational speed and performance. Code/models available here: https://piergiaj.github.io/rep-flow-site/

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

piergiaj/representation-flow-cvpr19 officialmentioned on GitHubpytorch report
KiUngSong/Vision mentioned on GitHubpytorch report
chunfeng0301/Flow mentioned on GitHubpaddle report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

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

Tasks

Action ClassificationAction RecognitionAction Recognition In VideosActivity RecognitionActivity Recognition In VideosOptical Flow EstimationTemporal Action LocalizationVideo ClassificationVideo Understanding

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 RepFlow-50 ([2+1]D CNN, FcF, Non-local block) Acc@1 77.9 #132 of 207 Archive leaderboard report
Action Recognition HMDB-51 RepFlow-50 ([2+1]D CNN, FcF, Non-local block) Average accuracy of 3 splits 81.1 #17 of 77 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

SPEED

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