AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Are Synthetic Data Useful for Egocentric Hand-Object Interaction Detection?

Rosario Leonardi, Antonino Furnari, Francesco Ragusa, Giovanni Maria Farinella

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Human-object interaction detection · 4 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 unlabeled real data via unsupervised domain adaptation to synthetic-only trainingVISOR HOS · EPIC-KITCHENS VISOR valOverall Mask AP: 09.88 -> 33.33Improved
add synthetic data and semi-supervised DA with 10% real labelsVISOR HOS · EPIC-KITCHENS VISOR val (10% labels)Overall Mask AP: 38.55 -> 44.22Improved
naively pre-train on synthetic data then fine-tune on 10% real data (no adaptation) (ablation)VISOR HOS · EPIC-KITCHENS VISOR val (10% labels)Overall Mask AP: 38.55 -> 37.62Hurt
add synthetic data with fully supervised domain adaptationVISOR HOS · EPIC-KITCHENS VISOR val (100% labels)Overall Mask AP: 45.33 -> 46.48Improved
add synthetic data with FSDAVISOR HOS · EgoHOS test (100% labels)Overall Mask AP: 36.16 -> 39.61Improved
add synthetic data with SSDA and 10% real labelsVISOR HOS · ENIGMA-51 test (10% labels)Overall Mask AP: 45.39 -> 57.08Improved
use in-domain synthetic data instead of out-domain synthetic data (unsupervised) (ablation)UDA on ENIGMA-51 · ENIGMA-51 test (0% labels)Overall Mask AP: 6.87 -> 34.78Improved
use in-domain instead of out-domain synthetic data with some real labels (ablation)SSDA on ENIGMA-51 · ENIGMA-51 test (10% labels)Overall Mask AP: 57.08 -> 56.69No clear effect
replace Adaptive Teacher with Mean Teacher, MT+GRL or Unbiased Teacher (ablation)SSDA · VISOR val (25% labels)Overall Mask AP: 45.55 -> 44.32Hurt
increase synthetic data amount up to about 22k-30k images (ablation)UDA on VISOR · VISOR valOverall Mask APImproved
use simulated hand-object images instead of pasting objects onto ImageNet backgrounds (ablation)UDA on VISOR · VISOR (3,500 image subset)Overall Mask AP: 19.62 -> 30.54Improved
use synthetic source instead of real VISOR data as UDA source for EgoHOS (ablation)UDA on EgoHOS · EgoHOS test (EPIC-excluded)Overall Mask AP: 24.39 -> 29.38Improved

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