Round Outcome Prediction in VALORANT Using Tactical Features from Video Analysis
This work addresses the problem of improving esports outcome prediction for VALORANT players and analysts, but it is incremental as it builds on existing video recognition methods.
The paper tackled predicting round outcomes in the FPS game VALORANT by analyzing tactical features from match footage, achieving approximately 81% accuracy from the middle phases onward, which significantly outperformed a baseline model using only minimap information.
Recently, research on predicting match outcomes in esports has been actively conducted, but much of it is based on match log data and statistical information. This research targets the FPS game VALORANT, which requires complex strategies, and aims to build a round outcome prediction model by analyzing minimap information in match footage. Specifically, based on the video recognition model TimeSformer, we attempt to improve prediction accuracy by incorporating detailed tactical features extracted from minimap information, such as character position information and other in-game events. This paper reports preliminary results showing that a model trained on a dataset augmented with such tactical event labels achieved approximately 81% prediction accuracy, especially from the middle phases of a round onward, significantly outperforming a model trained on a dataset with the minimap information itself. This suggests that leveraging tactical features from match footage is highly effective for predicting round outcomes in VALORANT.