Towards Cross-View Point Correspondence in Vision-Language Models

TL;DR

This paper introduces a new benchmark, dataset, and model for achieving precise cross-view point correspondence in vision-language models, addressing a key challenge in spatial understanding and embodied AI.

Contribution

The paper presents the CrossPoint-Bench benchmark, CrossPoint-378K dataset, and CroPond model, advancing the state-of-the-art in fine-grained cross-view correspondence in VLMs.

Findings

State-of-the-art models lag behind humans by over 54% in accuracy.

CroPond surpasses Gemini-2.5-Pro by 39.7% on the benchmark.

The dataset and benchmark facilitate progress in spatial understanding tasks.

Abstract

Cross-view correspondence is a fundamental capability for spatial understanding and embodied AI. However, it is still far from being realized in Vision-Language Models (VLMs), especially in achieving precise point-level correspondence, which is crucial for precise affordance interaction. So we propose the Cross-View Point Correspondence (CVPC) task and CrossPoint-Bench, a comprehensive benchmark with hierarchical design, inspired by the human cognitive process of "perceive", "reason", and "correspond". Our evaluation shows the state-of-the-art models (e.g., Gemini-2.5-Pro) still fall far behind humans, with a gap of over 54.65% in overall accuracy, exposing a challenge in transitioning from coarse-grained judgement to fine-grained coordinate prediction. To address this problem, we construct CrossPoint-378K, a dataset with 378K question-answering pairs across 900 scenes, focused on actionable affordance regions that better reflect real-world manipulation and interaction scenarios. Furthermore, we propose CroPond that trained on the CrossPoint-378K dataset. Our CroPond achieves state-of-the-art performance on CrossPoint-Bench, surpassing Gemini-2.5-Pro by 39.7% accuracy, which offers a foundation for advancing future work on cross-view correspondence. The benchmark, dataset, and model are publicly available at https://github.com/WangYipu2002/CrossPoint.