AI Paper Atlas

Research map / Video / Action recognition

Segmental Spatiotemporal CNNs for Fine-Grained Action Segmentation

Colin Lea, Austin Reiter, René Vidal, Gregory D. Hager

ECCV 2016 · Video › Action recognition · approach: Temporal action segmentation · 256 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 temporal 1D convolution over 10 s to the spatial CNN (ablation)S-CNN · 50 Salads (eval granularity)edit score: 24.10 -> 60.98Improved
add temporal 1D convolution over 10 s to the spatial CNN (ablation)S-CNN · 50 Salads (eval granularity)frame-wise accuracy: 66.64 -> 71.37Improved
add temporal 1D convolution to the spatial CNN (ablation)S-CNN · JIGSAWSframe-wise accuracy: 67.25 -> 71.21Improved
add segmental semi-Markov inference on top of ST-CNN (ablation)ST-CNN · JIGSAWSedit score: 59.89 -> 66.56Improved
add segmental semi-Markov inference on top of ST-CNN (ablation)ST-CNN · 50 Salads (eval granularity)edit score: 60.98 -> 62.06Improved
add segmental semi-Markov inference on top of ST-CNN (ablation)ST-CNN · JIGSAWSframe-wise accuracy: 71.21 -> 74.22Improved
use temporal CNN filters instead of LSTM on top of spatial featuresS-CNN + LSTM · JIGSAWSedit score: 54.07 -> 59.89Improved
use temporal CNN filters instead of LSTM on top of spatial featuresS-CNN + LSTM · 50 Salads (eval granularity)frame-wise accuracy: 66.30 -> 71.37Improved
add LSTM on top of spatial CNN (ablation)S-CNN · 50 Salads (eval granularity)frame-wise accuracy: 66.64 -> 66.30No clear effect
replace IDT features with the spatial CNNIDT · 50 Salads (eval granularity)frame-wise accuracy: 54.28 -> 66.64Improved
replace pre-trained VGG features with the task-trained spatial CNNVGG · 50 Salads (eval granularity)frame-wise accuracy: 38.30 -> 66.64Improved
use coarser action granularity (high, 4 classes instead of low, 52 classes)ST-CNN + Seg · 50 Saladsframe-wise accuracy: 44.13 -> 92.43Improved

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