Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Min, Junhong, Kim, Jimin, Kim, Minwook, Min, Cheol-Hui, Jeon, Youngpil, Choi, Minyong
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.16993
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915890311725056
author Min, Junhong
Kim, Jimin
Kim, Minwook
Min, Cheol-Hui
Jeon, Youngpil
Choi, Minyong
author_facet Min, Junhong
Kim, Jimin
Kim, Minwook
Min, Cheol-Hui
Jeon, Youngpil
Choi, Minyong
contents Depth in the real world is rarely singular. Transmissive materials create layered ambiguities that confound conventional perception systems. Existing models remain passive; conventional approaches typically estimate static depth maps anchored to the nearest surface, and even recent multi-head extensions suffer from a representational bottleneck due to fixed feature representations. This stands in contrast to human vision, which actively shifts focus to perceive a desired depth. We introduce \textbf{DepthFocus}, a steerable Vision Transformer that redefines stereo depth estimation as condition-aware control. Instead of extracting fixed features, our model dynamically modulates its computation based on a physical reference depth, integrating dual conditional mechanisms to selectively perceive geometry aligned with the desired focus. Leveraging a newly curated large-scale synthetic dataset, \textbf{DepthFocus} achieves state-of-the-art results across all evaluated benchmarks, including both standard single-layer and complex multi-layered scenarios. While maintaining high precision in opaque regions, our approach effectively resolves depth ambiguities in transparent and reflective scenes by selectively reconstructing geometry at a target distance. This capability enables robust, intent-driven perception that significantly outperforms existing multi-layer methods, marking a substantial step toward active 3D perception. \noindent \textbf{Project page}: \href{https://junhong-3dv.github.io/depthfocus-project/}{\textbf{this https URL}}.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16993
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DepthFocus: Controllable Depth Estimation for See-Through Scenes
Min, Junhong
Kim, Jimin
Kim, Minwook
Min, Cheol-Hui
Jeon, Youngpil
Choi, Minyong
Computer Vision and Pattern Recognition
Depth in the real world is rarely singular. Transmissive materials create layered ambiguities that confound conventional perception systems. Existing models remain passive; conventional approaches typically estimate static depth maps anchored to the nearest surface, and even recent multi-head extensions suffer from a representational bottleneck due to fixed feature representations. This stands in contrast to human vision, which actively shifts focus to perceive a desired depth. We introduce \textbf{DepthFocus}, a steerable Vision Transformer that redefines stereo depth estimation as condition-aware control. Instead of extracting fixed features, our model dynamically modulates its computation based on a physical reference depth, integrating dual conditional mechanisms to selectively perceive geometry aligned with the desired focus. Leveraging a newly curated large-scale synthetic dataset, \textbf{DepthFocus} achieves state-of-the-art results across all evaluated benchmarks, including both standard single-layer and complex multi-layered scenarios. While maintaining high precision in opaque regions, our approach effectively resolves depth ambiguities in transparent and reflective scenes by selectively reconstructing geometry at a target distance. This capability enables robust, intent-driven perception that significantly outperforms existing multi-layer methods, marking a substantial step toward active 3D perception. \noindent \textbf{Project page}: \href{https://junhong-3dv.github.io/depthfocus-project/}{\textbf{this https URL}}.
title DepthFocus: Controllable Depth Estimation for See-Through Scenes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.16993