DVD: Deterministic Video Depth Estimation with Generative Priors
2026-03-12 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present DVD, a new method that turns pre-trained video diffusion models into accurate depth predictors in one step. Their approach uses smart tricks to keep depth details sharp and consistent over time without needing lots of labeled data. They show that DVD works well right away on many tests, using much less data than other top methods. The authors also share their entire system openly for others to use.
video depth estimationvideo diffusion modelsdepth regressionlatent manifold rectificationaffine coherencezero-shot learninggeometric priorstemporal alignmentscale driftsemantic ambiguity
Authors
Hongfei Zhang, Harold Haodong Chen, Chenfei Liao, Jing He, Zixin Zhang, Haodong Li, Yihao Liang, Kanghao Chen, Bin Ren, Xu Zheng, Shuai Yang, Kun Zhou, Yinchuan Li, Nicu Sebe, Ying-Cong Chen
Abstract
Existing video depth estimation faces a fundamental trade-off: generative models suffer from stochastic geometric hallucinations and scale drift, while discriminative models demand massive labeled datasets to resolve semantic ambiguities. To break this impasse, we present DVD, the first framework to deterministically adapt pre-trained video diffusion models into single-pass depth regressors. Specifically, DVD features three core designs: (i) repurposing the diffusion timestep as a structural anchor to balance global stability with high-frequency details; (ii) latent manifold rectification (LMR) to mitigate regression-induced over-smoothing, enforcing differential constraints to restore sharp boundaries and coherent motion; and (iii) global affine coherence, an inherent property bounding inter-window divergence, which enables seamless long-video inference without requiring complex temporal alignment. Extensive experiments demonstrate that DVD achieves state-of-the-art zero-shot performance across benchmarks. Furthermore, DVD successfully unlocks the profound geometric priors implicit in video foundation models using 163x less task-specific data than leading baselines. Notably, we fully release our pipeline, providing the whole training suite for SOTA video depth estimation to benefit the open-source community.