Jamsrandorj, Ankhzaya Chao, Vanyi Nguyen, Hoang Quoc Oo, Yin May Robbani, Muhammad Amrulloh Hwang, Yewon Mun, Kyung Ryoul Kim, Jinwook 2025-10-31T01:30:09Z 2025-10-31T01:30:09Z 2025-10-23 2025-11-25 https://pubs.kist.re.kr/handle/201004/153379 Spatiotemporal event spotting in ball sports is essential for understanding complex game dynamics, requiring both high temporal precision and accurate spatial localization. However, most existing methods focus primarily on temporal localization, often neglecting the spatial dimensions that are crucial for tactical analysis. In this study, we propose a novel representation, 3D Heatmaps with Dynamically Shifted Gaussian Kernels, specifically designed to enable comprehensive spatiotemporal event spotting. To overcome current limitations, we introduce the Volleyball Nations League (VNL) dataset, which includes detailed annotations for eight key event types, encompassing both temporal and spatial labels. Our approach leverages a modified 3D U-Net architecture that effectively captures spatiotemporal patterns by utilizing our proposed heatmap design. Experimental results show that our method significantly outperforms state-of-the-art techniques in both temporal accuracy and spatial precision on the VNL dataset and a spatially augmented version of the SoccerNet Ball Action Spotting (BAS) dataset. These findings demonstrate the robustness and generalizability of our approach across different ball sports domains. English The British Machine Vision Association and Society for Pattern Recognition Spatiotemporal Event Spotting via 3D Heatmaps with Dynamically Shifted Gaussian Kernels Conference 1 The British Machine Vision Conference (BMVC) 2025 The British Machine Vision Conference (BMVC) 2025 UK Sheffield, UK 2025-11-24 Proceedings of The 36th British Machine Vision Conference