AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

A Novel Geometric Framework on Gram Matrix Trajectories for Human Behavior Understanding

Anis Kacem, Mohamed Daoudi, Boulbaba Ben Amor, Stefano Berretti, Juan Carlos Alvarez-Paiva

TPAMI 2018 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized sketching for matrices

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 spatial covariance (Gram matrix on S+) to Grassmann shape trajectories (ablation)landmark trajectory classifier · UT-Kinectaccuracy (%): 92.46 -> 96.48Improved
add spatial covariance (Gram matrix on S+) to Grassmann shape trajectories (ablation)landmark trajectory classifier · Florence3Daccuracy (%): 75 ± 5.22 -> 88.07 ± 4.8Improved
add spatial covariance (Gram matrix on S+) to Grassmann shape trajectories (ablation)landmark trajectory classifier · SBU Interactionaccuracy (%): 76.3 ± 3.26 -> 88.45 ± 2.88Improved
add spatial covariance (Gram matrix on S+) to Grassmann shape trajectories (ablation)landmark trajectory classifier · P-BME emotion recognitionaccuracy (%): 66.35 ± 6.43 -> 78.15 ± 5.79Improved
add dynamic time warping alignment (ablation)landmark trajectory classifier · UT-Kinectaccuracy (%): 91.46 -> 96.48Improved
add dynamic time warping alignment (ablation)landmark trajectory classifier · SBU Interactionaccuracy (%): 81.36 ± 2.78 -> 88.45 ± 2.88Improved
add dynamic time warping alignment (ablation)landmark trajectory classifier · P-BME emotion recognitionaccuracy (%): 63.23 ± 8.62 -> 78.15 ± 5.79Improved
use closeness on S+ instead of flat Frobenius distance (ablation)trajectory comparison · UT-Kinectaccuracy (%): 92.96 -> 96.48Improved
use closeness on S+ instead of regularized distance on positive definite cone (ablation)trajectory comparison · UT-Kinectaccuracy (%): 94.98 -> 96.48Improved
use ppfSVM instead of k-NN (ablation)S+ trajectory classifier · SBU Interactionaccuracy (%): 61.06 ± 2.3 -> 88.45 ± 2.88Improved
add late fusion of body-part classifiers (ablation)S+ trajectory classifier · SBU Interactionaccuracy (%): 88.45 ± 2.88 -> 93.7 ± 1.59Improved
add late fusion of body-part classifiers (ablation)S+ trajectory classifier · SYSU-3Daccuracy (%): 76.01 ± 2.09 -> 80.22 ± 2.09Improved

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