AI Paper Atlas

Research map / Large language models / AI and society

SALSA: A Novel Dataset for Multimodal Group Behavior Analysis

Xavier Alameda-Pineda, Jacopo Staiano, Ramanathan Subramanian, Ligia Batrinca, Elisa Ricci, Bruno Lepri, Oswald Lanz, Nicu Sebe

TPAMI 2016 · Large language models › AI and society · approach: Data commons and human baselines · 171 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 more training data (1% to 10%) for head posejoint head-body pose classifier · SALSA view 1 head posemean angular error (degrees): 45.7 -> 42.2Improved
use more training data (1% to 10%) for body posejoint head-body pose classifier · SALSA view 4 body posemean angular error (degrees): 54.6 -> 51.7Improved
estimate head instead of body posejoint head-body pose classifier · SALSA pose estimationmean angular error (degrees): 49.3 -> 45.7Improved
use body orientation instead of head orientationgraph-cut F-formation detection · SALSA F-formation detection (GT)F1: 0.82 -> 0.83Improved
use body orientation instead of head orientationHVFF ent · SALSA F-formation detection (GT)F1: 0.69 -> 0.73Improved
switch from Hough voting (HVFF lin) to graph cutF-formation detection · SALSA F-formation detection (GT head)F1: 0.63 -> 0.82Improved
concatenate MFCCs of all badges (frame-based fusion) instead of per-badge featuresspeaker recognition · SALSA speaker recognitionaccuracyImproved
add infra-red data to visual pose estimates (ablation)graph-cut F-formation detection · SALSA F-formation detection (estimated pose)F1Improved
add audio data to visual pose estimates (ablation)graph-cut F-formation detection · SALSA F-formation detection (estimated pose)F1No clear effect
use estimated visual pose instead of ground-truth posegraph-cut F-formation detection · SALSA F-formation detectionF1Hurt
move from Poster session to Party sessionHJS-PF tracker · SALSA trackingfailure rate (%): 2.6 -> 9.6Hurt
move from Poster session to Party sessionHJS-PF tracker · SALSA trackingprecision (cm): 15.2 -> 20.1Hurt

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