4D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillation

Despite advances in Multimodal LLMs (MLLMs), their ability to reason over 3D structures and temporal dynamics remains limited, constrained by weak 4D perception and temporal understanding. Existing 3D and 4D Video Question Answering (VQA) benchmarks also emphasize static scenes and lack region-level prompting. We tackle these issues by introducing: (a) 4D-RGPT, a specialized MLLM designed to capture 4D representations from video inputs with enhanced temporal perception; (b) Perceptual 4D Distillation (P4D), a training framework that transfers 4D representations from a frozen expert model into 4D-RGPT for comprehensive 4D perception; and (c) R4D-Bench, a benchmark for depth-aware dynamic scenes with region-level prompting, built via a hybrid automated and human-verified pipeline. Our 4D-RGPT achieves notable improvements on both existing 4D VQA benchmarks and the proposed R4D-Bench benchmark. Links: Project Page | GitHub | Dataset (a) Baseline R4D-Bench (b) Ours Ans: I am not sure.

0 2 12 ⟨R1⟩ Q: What is the average speed of ⟨R1⟩? Frame 1 Frame 2 Frame N () () Ans: 7.0 m/s Region (2D): Where is ⟨R1⟩? Depth (3D): How far away is ⟨R1⟩? Time (4D): When does ⟨R1⟩ move? Track ⟨R1⟩ across frames Depth Perception Temporal Perception Figure 1 | Overview of Region-level 4D Understanding. 4D region-level VQA, e.g., our R4D-Bench, requires MLLMs to be able to track regions (2D), perceive depth (3D), and temporal progression (4D).

Baseline MLLMs cannot recognize one or more of these aspects and thus fail to answer questions correctly. With our distillation framework, our 4D-RGPT better perceives these aspects and answers accurately. We note that the regions labeled with (*) are not provided in R4D-Bench; they are visualized for readability.