AI Paper Atlas

Research map / Video / Video-language understanding

Challenges in multimodal gesture recognition

Sergio Escalera, Vassilis Athitsos, Isabelle Guyon

JMLR 2016 · Video › Video-language understanding · approach: Egocentric video understanding · 0 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
use winning Motion Signature method instead of template-matching baselineBaseline method 1 · ChaLearn 2011-12 round 1, final set #1Levenshtein error rate: 0.6251 -> 0.0996Improved
use winning method instead of Principal Motion baselineBaseline method 2 · ChaLearn 2011-12 round 2, final set #1Levenshtein error rate: 0.2997 -> 0.0734Improved
move from round 1 to round 2 of the challenge (stronger baseline and entries)Alfnie · ChaLearn 2011-12 final set #1Levenshtein error rate: 0.0996 -> 0.0734Improved
add depth channel to RGB (Turtle Tamers vs Pennect)HOG/HOF HMM method · ChaLearn 2011-12 round 1/2Levenshtein error rate: 0.1652 -> 0.1702Mixed
use audio-based segmentation with audio+skeleton and late fusion (IVA MM) vs sliding-window skeleton-onlyJoewan-style skeleton sliding window · ChaLearn 2013 test setLevenshtein error rate: 0.83772 -> 0.12756Improved
evaluate on test set instead of validation setJOEWAN · ChaLearn 2013Levenshtein error rate: 0.13653 -> 0.83772Hurt
evaluate on test set instead of validation setWWEIGHT · ChaLearn 2013Levenshtein error rate: 0.46163 -> 0.15387Improved
add late fusion of audio and skeleton (WWEIGHT) compared with single-modality skeleton RFSUMO skeleton-only RF · ChaLearn 2013 test setLevenshtein error rate: 0.31652 -> 0.15387Improved

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