CuriosAI Submission to the EgoExo4D Proficiency Estimation Challenge 2025

TL;DR

This paper introduces two novel multi-view skill assessment methods for proficiency estimation, demonstrating improved accuracy through scenario-conditioned modeling techniques in the CVPR 2025 challenge.

Contribution

The paper presents a multi-task learning framework and a two-stage pipeline for proficiency estimation, advancing multi-view skill assessment methods.

Findings

Multi-task learning approach achieved 43.6% accuracy.

Two-stage pipeline achieved 47.8% accuracy.

Scenario-conditioned modeling improves proficiency estimation.

Abstract

This report presents the CuriosAI team's submission to the EgoExo4D Proficiency Estimation Challenge at CVPR 2025. We propose two methods for multi-view skill assessment: (1) a multi-task learning framework using Sapiens-2B that jointly predicts proficiency and scenario labels (43.6 % accuracy), and (2) a two-stage pipeline combining zero-shot scenario recognition with view-specific VideoMAE classifiers (47.8 % accuracy). The superior performance of the two-stage approach demonstrates the effectiveness of scenario-conditioned modeling for proficiency estimation.