Gameplay Video
Gameplay video analysis is a rapidly evolving field focused on extracting meaningful information and insights from video recordings of gameplay. Current research emphasizes developing AI models, often employing transformer architectures, convolutional neural networks, and recurrent neural networks, to perform tasks such as action recognition, pose estimation, emotion detection, and event segmentation within video sequences. These advancements are driving progress in areas like player experience modeling, video editing and generation, and even applications in healthcare (e.g., pain recognition and ADHD diagnosis) by leveraging the rich spatiotemporal data inherent in gameplay videos. The resulting techniques have significant implications for improving game design, enhancing user experience, and creating new possibilities for human-computer interaction.
Papers
DIFEM: Key-points Interaction based Feature Extraction Module for Violence Recognition in Videos
Himanshu Mittal, Suvramalya Basak, Anjali Gautam
LinVT: Empower Your Image-level Large Language Model to Understand Videos
Lishuai Gao, Yujie Zhong, Yingsen Zeng, Haoxian Tan, Dengjie Li, Zheng Zhao
Espresso: High Compression For Rich Extraction From Videos for Your Vision-Language Model
Keunwoo Peter Yu, Achal Dave, Rares Ambrus, Jean Mercat