ETA: Energy-based Test-time Adaptation for Depth Completion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chung, Younjoon, Park, Hyoungseob, Rim, Patrick, Zhang, Xiaoran, He, Jihe, Zeng, Ziyao, Cicek, Safa, Hong, Byung-Woo, Duncan, James S., Wong, Alex
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911111392002048
author Chung, Younjoon
Park, Hyoungseob
Rim, Patrick
Zhang, Xiaoran
He, Jihe
Zeng, Ziyao
Cicek, Safa
Hong, Byung-Woo
Duncan, James S.
Wong, Alex
author_facet Chung, Younjoon
Park, Hyoungseob
Rim, Patrick
Zhang, Xiaoran
He, Jihe
Zeng, Ziyao
Cicek, Safa
Hong, Byung-Woo
Duncan, James S.
Wong, Alex
contents We propose a method for test-time adaptation of pretrained depth completion models. Depth completion models, trained on some ``source'' data, often predict erroneous outputs when transferred to ``target'' data captured in novel environmental conditions due to a covariate shift. The crux of our method lies in quantifying the likelihood of depth predictions belonging to the source data distribution. The challenge is in the lack of access to out-of-distribution (target) data prior to deployment. Hence, rather than making assumptions regarding the target distribution, we utilize adversarial perturbations as a mechanism to explore the data space. This enables us to train an energy model that scores local regions of depth predictions as in- or out-of-distribution. We update the parameters of pretrained depth completion models at test time to minimize energy, effectively aligning test-time predictions to those of the source distribution. We call our method ``Energy-based Test-time Adaptation'', or ETA for short. We evaluate our method across three indoor and three outdoor datasets, where ETA improve over the previous state-of-the-art method by an average of 6.94% for outdoors and 10.23% for indoors. Project Page: https://fuzzythecat.github.io/eta.
format Preprint
id arxiv_https___arxiv_org_abs_2508_05989
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ETA: Energy-based Test-time Adaptation for Depth Completion
Chung, Younjoon
Park, Hyoungseob
Rim, Patrick
Zhang, Xiaoran
He, Jihe
Zeng, Ziyao
Cicek, Safa
Hong, Byung-Woo
Duncan, James S.
Wong, Alex
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
We propose a method for test-time adaptation of pretrained depth completion models. Depth completion models, trained on some ``source'' data, often predict erroneous outputs when transferred to ``target'' data captured in novel environmental conditions due to a covariate shift. The crux of our method lies in quantifying the likelihood of depth predictions belonging to the source data distribution. The challenge is in the lack of access to out-of-distribution (target) data prior to deployment. Hence, rather than making assumptions regarding the target distribution, we utilize adversarial perturbations as a mechanism to explore the data space. This enables us to train an energy model that scores local regions of depth predictions as in- or out-of-distribution. We update the parameters of pretrained depth completion models at test time to minimize energy, effectively aligning test-time predictions to those of the source distribution. We call our method ``Energy-based Test-time Adaptation'', or ETA for short. We evaluate our method across three indoor and three outdoor datasets, where ETA improve over the previous state-of-the-art method by an average of 6.94% for outdoors and 10.23% for indoors. Project Page: https://fuzzythecat.github.io/eta.
title ETA: Energy-based Test-time Adaptation for Depth Completion
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2508.05989