DepthEvidence: Unifying Metric Depth Prediction and Geometric Reasoning in Multimodal Language Models

arXiv:2609.31103v1 Announce Type: cross Abstract: Spatial reasoning with metric constraints requires linking objects to geometric measurements and preserving their numerical content during language reasoning. We present DepthEvidence, a 4B model that uses its own dense metric predictions as object-grounded evidence for language generation. A camera-conditioned decoder predicts full-resolution metric depth using multi-scale visual features and high-resolution RGB refinement. A dense-to-language interface converts predicted depths and decoder features into object-aligned continuous geometry tokens anchored to object identifiers. Geometric supervision encourages metric information to remain recoverable before and after language-context interaction, while instruction tuning supports object measurement and compositional reasoning. We introduce a Depth-VQA benchmark evaluating object-depth queries, relative comparisons, and decisions combining spatial and numerical constraints. Across nine datasets, DepthEvidence achieves the highest average dense elta_1 among evaluated methods, competitive with specialized estimators. It also leads the evaluated methods in instance-level metric depth estimation and overall accuracy on both relative and metric reasoning tracks, while broadly preserving general VQA performance and improving spatial understanding relative to the base model.

Related

• CrossView Suite: Harnessing Cross-view Spatial Intelligence of MLLMs with Dataset, Model and Benchmark
• CapGeo-Bench: Decoupling Visual Perception from Reasoning and Evaluating Geometric Understanding
• From Symbolic to Geometric: Enabling Spatial Reasoning in Large Language Models
• Symbolic Intermediaries as a Linguistic-Numerical Interface for LLM-Driven Geometric Reasoning

Source: arXiv cs.AI | 2026-09-28