EgoEsportsQA: An Egocentric Video Benchmark for Perception and Reasoning in Esports

TL;DR

EgoEsportsQA introduces a new benchmark dataset for evaluating perception and reasoning in egocentric esports videos, highlighting current model limitations in high-velocity virtual environments.

Contribution

The paper presents a novel QA benchmark with 1,745 questions from professional esports matches, structured into a taxonomy to evaluate perception and reasoning capabilities of Video-LLMs.

Findings

Current Video-LLMs achieve only 71.58% accuracy on the benchmark.

Models perform better in perception than in tactical reasoning.

Deep micro-operations remain challenging for existing models.

Abstract

While video large language models (Video-LLMs) excel in understanding slow-paced, real-world egocentric videos, their capabilities in high-velocity, information-dense virtual environments remain under-explored. Existing benchmarks focus on daily activities, yet lack a rigorous testbed for evaluating fast, rule-bound reasoning in virtual scenarios. To fill this gap, we introduce EgoEsportsQA, a pioneering video question-answering (QA) benchmark for grounding perception and reasoning in expert esports knowledge. We curate 1,745 high-quality QA pairs from professional matches across 3 first-person shooter games via a scalable six-stage pipeline. These questions are structured into a two-dimensional decoupled taxonomy: 11 sub-tasks in the cognitive capability dimension (covering perception and reasoning levels) and 6 sub-tasks in the esports knowledge dimension. Comprehensive evaluations of state-of-the-art Video-LLMs reveal that current models still fail to achieve satisfactory performance, with the best model only 71.58%. The results expose notable gaps across both axes: models exhibit stronger capabilities in basic visual perception than in deep tactical reasoning, and they grasp overall macro-progression better than fine-grained micro-operations. Extensive ablation experiments demonstrate the intrinsic weaknesses of current Video-LLM architectures. Further analysis suggests that our dataset not only reveals the connections between real-world and virtual egocentric domains, but also offers guidance for optimizing downstream esports applications, thereby fostering the future advancement of Video-LLMs in various egocentric environments.