AI Paper Atlas

Research map / Video / Action recognition

FinePseudo: Improving Pseudo-Labelling through Temporal-Alignablity for Semi-Supervised Fine-Grained Action Recognition

Ishan Rajendrakumar Dave, Mamshad Nayeem Rizve, Mubarak Shah

ECCV 2024 · Video › Action recognition · approach: Temporal action segmentation · 5 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 FinePseudo collaborative pseudo-labelingR2plus1D-18 · Diving48 (10% labels)top-1 accuracy: 33.7 -> 37.6Improved
add FinePseudo to AIM CLIP ViT-BAIM ViT-B · Diving48 (5% labels)top-1 accuracy: 38.12 -> 43.02Improved
add FinePseudo in open-world setting with novel classesR2plus1D-18 · Diving48 open-world (10% labels)top-1 accuracy: 39.60 -> 42.21Improved
use regular pseudo-labeling in open-world settingR2plus1D-18 · Diving48 open-world (10% labels)top-1 accuracy: 39.60 -> 38.29Hurt
add alignability-verification metric learning (both losses) to alignability encoder (ablation)FinePseudo · Diving48 (10% labels)top-1 accuracy: 33.50 -> 37.64Improved
add triplet loss L_AT in addition to score loss (ablation)FinePseudo · Diving48 (10% labels)top-1 accuracy: 36.11 -> 37.64Improved
add GITDL self-supervised pretraining for alignability encoder (ablation)FinePseudo · Diving48 (10% labels)top-1 accuracy: 35.23 -> 37.64Improved
combine alignability encoder with action encoder rather than using either alone (ablation)FinePseudo · Diving48 (10% labels)top-1 accuracy: 33.10 -> 37.64Improved
add uncertainty filtering to confidence-based pseudo-labels (ablation)pseudo-labeling · Diving48 (10% labels)top-1 accuracy: 33.40 -> 33.58No clear effect
verify pseudo-labels with alignability score (ablation)pseudo-labeling · Diving48 (10% labels)top-1 accuracy: 33.40 -> 37.09Improved
use non-parametric alignability classifier for pseudo-labels instead of verification only (ablation)pseudo-labeling · Diving48 (10% labels)top-1 accuracy: 37.09 -> 37.64Improved
use hard negatives instead of all negatives in triplet mining (ablation)alignability metric learning · Diving48 (10% labels)top-1 accuracy: 36.16 -> 37.64Improved

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