Learning A Zero-shot Occupancy Network from Vision Foundation Models via Self-supervised Adaptation

• URL: http://arxiv.org/abs/2503.07125v1
• Date: Mon, 10 Mar 2025 09:54:40 GMT
• Title: Learning A Zero-shot Occupancy Network from Vision Foundation Models via Self-supervised Adaptation
• Authors: Sihao Lin, Daqi Liu, Ruochong Fu, Dongrui Liu, Andy Song, Hongwei Xie, Zhihui Li, Bing Wang, Xiaojun Chang,
• Abstract summary: This work proposes a novel approach that bridges 2D vision foundation models with 3D tasks.<n>We leverage the zero-shot capabilities of vision-language models for image semantics.<n>We project the semantics into 3D space using the reconstructed metric depth, thereby providing 3D supervision.
• Score: 41.98740330990215
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Estimating the 3D world from 2D monocular images is a fundamental yet challenging task due to the labour-intensive nature of 3D annotations. To simplify label acquisition, this work proposes a novel approach that bridges 2D vision foundation models (VFMs) with 3D tasks by decoupling 3D supervision into an ensemble of image-level primitives, e.g., semantic and geometric components. As a key motivator, we leverage the zero-shot capabilities of vision-language models for image semantics. However, due to the notorious ill-posed problem - multiple distinct 3D scenes can produce identical 2D projections, directly inferring metric depth from a monocular image in a zero-shot manner is unsuitable. In contrast, 2D VFMs provide promising sources of relative depth, which theoretically aligns with metric depth when properly scaled and offset. Thus, we adapt the relative depth derived from VFMs into metric depth by optimising the scale and offset using temporal consistency, also known as novel view synthesis, without access to ground-truth metric depth. Consequently, we project the semantics into 3D space using the reconstructed metric depth, thereby providing 3D supervision. Extensive experiments on nuScenes and SemanticKITTI demonstrate the effectiveness of our framework. For instance, the proposed method surpasses the current state-of-the-art by 3.34% mIoU on nuScenes for voxel occupancy prediction.

Related papers

• MOSE: Monocular Semantic Reconstruction Using NeRF-Lifted Noisy Priors [11.118490283303407]
  We propose a neural field semantic reconstruction approach to lift inferred image-level noisy priors to 3D. Our method produces accurate semantics and geometry in both 3D and 2D space.
  arXiv  Detail & Related papers  (2024-09-21T05:12:13Z)
• GEOcc: Geometrically Enhanced 3D Occupancy Network with Implicit-Explicit Depth Fusion and Contextual Self-Supervision [49.839374549646884]
  This paper presents GEOcc, a Geometric-Enhanced Occupancy network tailored for vision-only surround-view perception. Our approach achieves State-Of-The-Art performance on the Occ3D-nuScenes dataset with the least image resolution needed and the most weightless image backbone.
  arXiv  Detail & Related papers  (2024-05-17T07:31:20Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [114.47216525866435]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• SSR-2D: Semantic 3D Scene Reconstruction from 2D Images [54.46126685716471]
  In this work, we explore a central 3D scene modeling task, namely, semantic scene reconstruction without using any 3D annotations. The key idea of our approach is to design a trainable model that employs both incomplete 3D reconstructions and their corresponding source RGB-D images. Our method achieves the state-of-the-art performance of semantic scene completion on two large-scale benchmark datasets MatterPort3D and ScanNet.
  arXiv  Detail & Related papers  (2023-02-07T17:47:52Z)
• Monocular 3D Object Detection with Depth from Motion [74.29588921594853]
  We take advantage of camera ego-motion for accurate object depth estimation and detection. Our framework, named Depth from Motion (DfM), then uses the established geometry to lift 2D image features to the 3D space and detects 3D objects thereon. Our framework outperforms state-of-the-art methods by a large margin on the KITTI benchmark.
  arXiv  Detail & Related papers  (2022-07-26T15:48:46Z)
• Learning Ego 3D Representation as Ray Tracing [42.400505280851114]
  We present a novel end-to-end architecture for ego 3D representation learning from unconstrained camera views. Inspired by the ray tracing principle, we design a polarized grid of "imaginary eyes" as the learnable ego 3D representation. We show that our model outperforms all state-of-the-art alternatives significantly.
  arXiv  Detail & Related papers  (2022-06-08T17:55:50Z)
• Learning Geometry-Guided Depth via Projective Modeling for Monocular 3D Object Detection [70.71934539556916]
  We learn geometry-guided depth estimation with projective modeling to advance monocular 3D object detection. Specifically, a principled geometry formula with projective modeling of 2D and 3D depth predictions in the monocular 3D object detection network is devised. Our method remarkably improves the detection performance of the state-of-the-art monocular-based method without extra data by 2.80% on the moderate test setting.
  arXiv  Detail & Related papers  (2021-07-29T12:30:39Z)
• MonoGRNet: A General Framework for Monocular 3D Object Detection [23.59839921644492]
  We propose MonoGRNet for the amodal 3D object detection from a monocular image via geometric reasoning. MonoGRNet decomposes the monocular 3D object detection task into four sub-tasks including 2D object detection, instance-level depth estimation, projected 3D center estimation and local corner regression. Experiments are conducted on KITTI, Cityscapes and MS COCO datasets.
  arXiv  Detail & Related papers  (2021-04-18T10:07:52Z)
• Monocular Differentiable Rendering for Self-Supervised 3D Object Detection [21.825158925459732]
  3D object detection from monocular images is an ill-posed problem due to the projective entanglement of depth and scale. We present a novel self-supervised method for textured 3D shape reconstruction and pose estimation of rigid objects. Our method predicts the 3D location and meshes of each object in an image using differentiable rendering and a self-supervised objective.
  arXiv  Detail & Related papers  (2020-09-30T09:21:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.