PhysScene: A Scene Graph Dataset for Scientific Visual Reasoning in Physics Experiments

2026-06-08

AI Summary

Existing scene graph datasets are predominantly confined to generic natural scenes, limiting their applicability to structured visual reasoning in scientific experimental settings. To address this gap, this work introduces PhysScene, the first scene graph dataset specifically designed for physics experiments, focusing on specialized instruments and experimental apparatus. PhysScene annotates functional and logical dependencies among objects, emphasizing semantic constraints and high relational density. Departing from conventional scale-driven data collection paradigms, PhysScene establishes a new benchmark for scientific visual reasoning, challenges state-of-the-art scene parsing algorithms, and provides a high-quality, publicly available evaluation platform.

Abstract

Scene Graphs (SGs) provide structured representations of visual scenes by modeling objects and their pairwise relationships. Despite recent progress, existing datasets primarily focus on generic natural contexts, leaving domain-specific and function-oriented scenes largely underexplored. This limitation restricts the evaluation of relational reasoning in scientific experimental scenes, thereby hindering the development of intelligent monitoring, analysis, and related applications in such scenes. To address this gap, we introduce PhysScene, the first SG dataset tailored to physics experiments. PhysScene encompasses specialized instruments, structured experimental setups, and functional relations intrinsic to experimental environments, enabling reasoning that extends beyond spatial co-occurrence to logical dependencies. Rather than pursuing large data scale, PhysScene focuses on strong semantic constraints and high relation density in experimental scenes, posing new challenges for existing scene parsing algorithms while offering opportunities for further improvements. Extensive analyses and experiments show that PhysScene complements existing benchmarks and establishes a valuable testbed for advancing scientific visual reasoning. The dataset is publicly available at https://github.com/ZMH-SDUST/PhysScene.

Problem

• Scene Graphs
• Scientific Visual Reasoning
• Physics Experiments
• Relational Reasoning
• Domain-specific Scenes

Innovation

• Scene Graph
• Scientific Visual Reasoning
• Physics Experiments
• Functional Relations
• Semantic Constraints