RAVU builds a spatio-temporal graph memory of a video and uses compositional reasoning over the graph to retrieve a few key frames, achieving state-of-the-art zero-shot video QA on NExT-QA and EgoSchema with 5-10 frames.
Multi-object event graph representation learning for Video Question Answering
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Video question answering (VideoQA) is a task to predict the correct answer to questions posed about a given video. The system must comprehend spatial and temporal relationships among objects extracted from videos to perform causal and temporal reasoning. While prior works have focused on modeling individual object movements using transformer-based methods, they falter when capturing complex scenarios involving multiple objects (e.g., "a boy is throwing a ball in a hoop"). We propose a contrastive language event graph representation learning method called CLanG to address this limitation. Aiming to capture event representations associated with multiple objects, our method employs a multi-layer GNN-cluster module for adversarial graph representation learning, enabling contrastive learning between the question text and its relevant multi-object event graph. Our method outperforms a strong baseline, achieving up to 2.2% higher accuracy on two challenging VideoQA datasets, NExT-QA and TGIF-QA-R. In particular, it is 2.8% better than baselines in handling causal and temporal questions, highlighting its strength in reasoning multiple object-based events.
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
RAVU: Retrieval Augmented Video Understanding with Compositional Reasoning over Graph
RAVU builds a spatio-temporal graph memory of a video and uses compositional reasoning over the graph to retrieve a few key frames, achieving state-of-the-art zero-shot video QA on NExT-QA and EgoSchema with 5-10 frames.