Research

Paper

AI LLM March 02, 2026

Efficient Test-Time Optimization for Depth Completion via Low-Rank Decoder Adaptation

Authors

Minseok Seo, Wonjun Lee, Jaehyuk Jang, Changick Kim

Abstract

Zero-shot depth completion has gained attention for its ability to generalize across environments without sensor-specific datasets or retraining. However, most existing approaches rely on diffusion-based test-time optimization, which is computationally expensive due to iterative denoising. Recent visual-prompt-based methods reduce training cost but still require repeated forward--backward passes through the full frozen network to optimize input-level prompts, resulting in slow inference. In this work, we show that adapting only the decoder is sufficient for effective test-time optimization, as depth foundation models concentrate depth-relevant information within a low-dimensional decoder subspace. Based on this insight, we propose a lightweight test-time adaptation method that updates only this low-dimensional subspace using sparse depth supervision. Our approach achieves state-of-the-art performance, establishing a new Pareto frontier between accuracy and efficiency for test-time adaptation. Extensive experiments on five indoor and outdoor datasets demonstrate consistent improvements over prior methods, highlighting the practicality of fast zero-shot depth completion.

Metadata

arXiv ID: 2603.01765
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-02
Fetched: 2026-03-03 04:34

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.01765v1</id>\n    <title>Efficient Test-Time Optimization for Depth Completion via Low-Rank Decoder Adaptation</title>\n    <updated>2026-03-02T11:45:19Z</updated>\n    <link href='https://arxiv.org/abs/2603.01765v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.01765v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Zero-shot depth completion has gained attention for its ability to generalize across environments without sensor-specific datasets or retraining. However, most existing approaches rely on diffusion-based test-time optimization, which is computationally expensive due to iterative denoising. Recent visual-prompt-based methods reduce training cost but still require repeated forward--backward passes through the full frozen network to optimize input-level prompts, resulting in slow inference. In this work, we show that adapting only the decoder is sufficient for effective test-time optimization, as depth foundation models concentrate depth-relevant information within a low-dimensional decoder subspace. Based on this insight, we propose a lightweight test-time adaptation method that updates only this low-dimensional subspace using sparse depth supervision. Our approach achieves state-of-the-art performance, establishing a new Pareto frontier between accuracy and efficiency for test-time adaptation. Extensive experiments on five indoor and outdoor datasets demonstrate consistent improvements over prior methods, highlighting the practicality of fast zero-shot depth completion.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-02T11:45:19Z</published>\n    <arxiv:comment>17 pages, 7 figures [We achieved a new Pareto frontier in test-time depth completion.]</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Minseok Seo</name>\n    </author>\n    <author>\n      <name>Wonjun Lee</name>\n    </author>\n    <author>\n      <name>Jaehyuk Jang</name>\n    </author>\n    <author>\n      <name>Changick Kim</name>\n    </author>\n  </entry>"
}