AI Paper Atlas

Research map / Video / Action recognition

R-C3D: Region Convolutional 3D Network for Temporal Activity Detection

Huijuan Xu, Abir Das, Kate Saenko

ICCV 2017 · Video › Action recognition · approach: Fully supervised temporal action detection · 793 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the reverse pass (two-way buffer) for training data (ablation)R-C3D · THUMOS'14mAP@0.5 (%): 27.0 -> 28.9Improved
add the reverse pass (two-way buffer) for training data (ablation)R-C3D · THUMOS'14 mAP@0.1mAP@0.5 (%): 51.6 -> 54.5Improved
use end-to-end R-C3D instead of best prior methodR-C3D · THUMOS'14mAP@0.5 (%): 23.3 -> 28.9Improved
use R-C3D instead of UPC (C3D features only)R-C3D · ActivityNet validationmAP@0.5 (%): 22.5 -> 26.8Improved
use R-C3D instead of UPC (C3D features only)R-C3D · ActivityNet testmAP@0.5 (%): 22.3 -> 26.8Improved
train on train+val instead of train only (ablation)R-C3D · ActivityNet testmAP@0.5 (%): 26.8 -> 28.4Improved
train on train+val instead of train only (ablation)R-C3D · ActivityNet test average mAPaverage mAP (%): 13.1 -> 16.7Improved
use R-C3D instead of methods with handcrafted featuresR-C3D · ActivityNet testmAP@0.5 (%): 28.4 vs 36.4Hurt
use R-C3D instead of asynchronous temporal fields (standard evaluation)R-C3D · CharadesmAP (%): 9.6 -> 12.4Improved
use R-C3D instead of asynchronous temporal fields (post-processed evaluation)R-C3D · CharadesmAP (%): 12.1 -> 12.7Mixed
vary anchor segment scales (ablation)R-C3D · CharadesmAP (%)No clear effect
share conv features between proposal and classification subnets, versus LSTM-based DAPR-C3D · inference speedFPS: 134.1 -> 569 FPSImproved

Later papers that compared with this method

9 later accepted papers report R-C3D as a baseline in their results tables; 8 of them report a better number on most of the results they share with it.

1 more not listed.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy