AI Paper Atlas

Research map / Video / Action recognition

Learning Motion in Feature Space: Locally-Consistent Deformable Convolution Networks for Fine-Grained Action Detection

Khoi-Nguyen C. Mac, Dhiraj Joshi, Raymond A. Yeh, Jinjun Xiong, Rogerio S. Feris, Minh N. Do

ICCV 2019 · Video › Action recognition · approach: Fully supervised temporal action detection · 47 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 local coherency constraint to deformable convolution (ablation)LCDC · 50 Salads (split 1, mid-level, ablation)frame-wise accuracy: 72.25 -> 73.77Improved
add the local coherency constraint (deformation parameters) (ablation)LCDC · 50 Salads (split 1, mid-level, ablation)deformable parameters: 995.5K -> 27.7KImproved
use LCDC features instead of SpatialCNN with ED-TCNED-TCN · 50 Salads (mid-level)F1@10: 68.0 -> 73.75Improved
use LCDC features instead of SpatialCNN with ED-TCNED-TCN · 50 Salads (eval-level)frame-wise accuracy: 73.4 -> 78.90Improved
use LCDC features instead of SpatialCNN with ED-TCNED-TCN · GTEAF1@10: 72.2 -> 75.39Improved
use LCDC features instead of SpatialCNN with ED-TCNED-TCN · GTEAframe-wise accuracy: 64.0 -> 65.34Improved
use LCDC features alone instead of SpatialCNN featuresLCDC · 50 Salads (mid-level)frame-wise accuracy: 54.9 -> 67.27Improved
add multiple averaged feature frames to appearance stream (ablation)NaiveAppear · 50 Salads (split 1, mid-level, ablation)frame-wise accuracy: 68.45 -> 71.52Improved
replace two-stream with optical flow by learned deformation (LCDC) (ablation)TwoStreamNet · 50 Salads (split 1, mid-level, ablation)frame-wise accuracy: 71.82 -> 73.77Improved
use LCDC features instead of SpatialCNN features at GTEA, no long-temporal modelLCDC · GTEAframe-wise accuracy: 54.1 -> 55.32Mixed
use LCDC features with ST-CNN rather than ED-TCNLCDC · 50 Salads (mid-level)F1@10: 73.75 -> 60.01Hurt

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