AI Paper Atlas

Research map / Video / Action recognition

Leveraging temporal contextualization for video action recognition

Minji Kim, Dongyoon Han, Taekyung Kim, Bohyung Han

ECCV 2024 · Video › Action recognition · approach: Contrastive video representation learning · 19 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 Contextualization (TC) to ViFi-CLIP baseline (ablation)ViFi-CLIP baseline · Zero-shot (no WE)top-1 accuracy (All): 67.3 -> 68.0Improved
add Video-conditional Prompting (VP) (ablation)ViFi-CLIP baseline · Zero-shot (no WE)top-1 accuracy (All): 67.3 -> 68.4Improved
add TC and VP with weight-space ensembling (ablation)ViFi-CLIP baseline · Zero-shot (WE)top-1 accuracy (All): 69.0 -> 71.0Improved
use all patch tokens from all frames as reference (joint space-time) (ablation)frame-wise baseline · HMDB-51 16-shotaccuracy: 67.1 -> 63.3Hurt
use context tokens instead of CLS or adjacent-frame tokens as reference (ablation)frame-wise baseline · SSv2 16-shotaccuracy: 12.0 -> 14.3Improved
use seed tokens without merging into context tokens (ablation)TC · Few-shot averagetop-1 accuracy (All): 53.5 -> 50.2Hurt
merge tokens randomly rather than by relevance (ablation)TC · Few-shot averagetop-1 accuracy (All): 53.5 -> 51.2Hurt
select seed tokens by averaged CLS attention instead of no selection (ablation)TC · Few-shot averagetop-1 accuracy (All): 54.1 -> 54.5Improved
vary seed token ratio alpha from 0.2 to 0.6 (ablation)TC · Few-shot averagetop-1 accuracy (All): 54.2 to 54.5No clear effect
add learnable bias between local and global keys with spatial embedding (ablation)TC · Few-shot averagetop-1 accuracy (All): 54.2 -> 54.5Improved
add learnable prompt vectors without video conditioning (ablation)ViFi-CLIP baseline · Zero-shot (no WE)top-1 accuracy (All): 67.3 -> 67.1Hurt
use VP with context tokens instead of vision-text late-fusion (ablation)VP · Zero-shot K-600top-1 accuracy: 70.9 -> 72.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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