MonoVQD: Monocular 3D Object Detection with Variational Query Denoising and Self-Distillation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Vu, Kiet Dang, Tran, Trung Thai, Nguyen, Duc Dung
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911010570371072
author Vu, Kiet Dang
Tran, Trung Thai
Nguyen, Duc Dung
author_facet Vu, Kiet Dang
Tran, Trung Thai
Nguyen, Duc Dung
contents Precisely localizing 3D objects from a single image constitutes a central challenge in monocular 3D detection. While DETR-like architectures offer a powerful paradigm, their direct application in this domain encounters inherent limitations, preventing optimal performance. Our work addresses these challenges by introducing MonoVQD, a novel framework designed to fundamentally advance DETR-based monocular 3D detection. We propose three main contributions. First, we propose the Mask Separated Self-Attention mechanism that enables the integration of the denoising process into a DETR architecture. This improves the stability of Hungarian matching to achieve a consistent optimization objective. Second, we present the Variational Query Denoising technique to address the gradient vanishing problem of conventional denoising methods, which severely restricts the efficiency of the denoising process. This explicitly introduces stochastic properties to mitigate this fundamental limitation and unlock substantial performance gains. Finally, we introduce a sophisticated self-distillation strategy, leveraging insights from later decoder layers to synergistically improve query quality in earlier layers, thereby amplifying the iterative refinement process. Rigorous experimentation demonstrates that MonoVQD achieves superior performance on the challenging KITTI monocular benchmark. Highlighting its broad applicability, MonoVQD's core components seamlessly integrate into other architectures, delivering significant performance gains even in multi-view 3D detection scenarios on the nuScenes dataset and underscoring its robust generalization capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14835
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MonoVQD: Monocular 3D Object Detection with Variational Query Denoising and Self-Distillation
Vu, Kiet Dang
Tran, Trung Thai
Nguyen, Duc Dung
Computer Vision and Pattern Recognition
Precisely localizing 3D objects from a single image constitutes a central challenge in monocular 3D detection. While DETR-like architectures offer a powerful paradigm, their direct application in this domain encounters inherent limitations, preventing optimal performance. Our work addresses these challenges by introducing MonoVQD, a novel framework designed to fundamentally advance DETR-based monocular 3D detection. We propose three main contributions. First, we propose the Mask Separated Self-Attention mechanism that enables the integration of the denoising process into a DETR architecture. This improves the stability of Hungarian matching to achieve a consistent optimization objective. Second, we present the Variational Query Denoising technique to address the gradient vanishing problem of conventional denoising methods, which severely restricts the efficiency of the denoising process. This explicitly introduces stochastic properties to mitigate this fundamental limitation and unlock substantial performance gains. Finally, we introduce a sophisticated self-distillation strategy, leveraging insights from later decoder layers to synergistically improve query quality in earlier layers, thereby amplifying the iterative refinement process. Rigorous experimentation demonstrates that MonoVQD achieves superior performance on the challenging KITTI monocular benchmark. Highlighting its broad applicability, MonoVQD's core components seamlessly integrate into other architectures, delivering significant performance gains even in multi-view 3D detection scenarios on the nuScenes dataset and underscoring its robust generalization capabilities.
title MonoVQD: Monocular 3D Object Detection with Variational Query Denoising and Self-Distillation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.14835