DiffCAD: Weakly-Supervised Probabilistic CAD Model Retrieval and Alignment from an RGB Image

• URL: http://arxiv.org/abs/2311.18610v2
• Date: Thu, 6 Jun 2024 14:37:01 GMT
• Title: DiffCAD: Weakly-Supervised Probabilistic CAD Model Retrieval and Alignment from an RGB Image
• Authors: Daoyi Gao, Dávid Rozenberszki, Stefan Leutenegger, Angela Dai,
• Abstract summary: We propose DiffCAD, the first weakly-supervised probabilistic approach to CAD retrieval and alignment from an RGB image. We formulate this as a conditional generative task, leveraging diffusion to learn implicit probabilistic models capturing the shape, pose, and scale of CAD objects in an image. Our approach is trained only on synthetic data, leveraging monocular depth and mask estimates to enable robust zero-shot adaptation to various real target domains.
• Score: 34.47379913018661
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Perceiving 3D structures from RGB images based on CAD model primitives can enable an effective, efficient 3D object-based representation of scenes. However, current approaches rely on supervision from expensive annotations of CAD models associated with real images, and encounter challenges due to the inherent ambiguities in the task -- both in depth-scale ambiguity in monocular perception, as well as inexact matches of CAD database models to real observations. We thus propose DiffCAD, the first weakly-supervised probabilistic approach to CAD retrieval and alignment from an RGB image. We formulate this as a conditional generative task, leveraging diffusion to learn implicit probabilistic models capturing the shape, pose, and scale of CAD objects in an image. This enables multi-hypothesis generation of different plausible CAD reconstructions, requiring only a few hypotheses to characterize ambiguities in depth/scale and inexact shape matches. Our approach is trained only on synthetic data, leveraging monocular depth and mask estimates to enable robust zero-shot adaptation to various real target domains. Despite being trained solely on synthetic data, our multi-hypothesis approach can even surpass the supervised state-of-the-art on the Scan2CAD dataset by 5.9% with 8 hypotheses.

Related papers

• Towards Human-Level 3D Relative Pose Estimation: Generalizable, Training-Free, with Single Reference [62.99706119370521]
  Humans can easily deduce the relative pose of an unseen object, without label/training, given only a single query-reference image pair. We propose a novel 3D generalizable relative pose estimation method by elaborating (i) with a 2.5D shape from an RGB-D reference, (ii) with an off-the-shelf differentiable, and (iii) with semantic cues from a pretrained model like DINOv2.
  arXiv  Detail & Related papers  (2024-06-26T16:01:10Z)
• Sparse Multi-Object Render-and-Compare [33.97243145891282]
  Reconstructing 3D shape and pose of static objects from a single image is an essential task for various industries. Directly predicting 3D shapes produces unrealistic, overly smoothed or tessellated shapes. Retrieving CAD models ensures realistic shapes but requires robust and accurate alignment.
  arXiv  Detail & Related papers  (2023-10-17T12:01:32Z)
• Model2Scene: Learning 3D Scene Representation via Contrastive Language-CAD Models Pre-training [105.3421541518582]
  Current successful methods of 3D scene perception rely on the large-scale annotated point cloud. We propose Model2Scene, a novel paradigm that learns free 3D scene representation from Computer-Aided Design (CAD) models and languages. Model2Scene yields impressive label-free 3D object salient detection with an average mAP of 46.08% and 55.49% on the ScanNet and S3DIS datasets, respectively.
  arXiv  Detail & Related papers  (2023-09-29T03:51:26Z)
• MegaPose: 6D Pose Estimation of Novel Objects via Render & Compare [84.80956484848505]
  MegaPose is a method to estimate the 6D pose of novel objects, that is, objects unseen during training. We present a 6D pose refiner based on a render&compare strategy which can be applied to novel objects. Second, we introduce a novel approach for coarse pose estimation which leverages a network trained to classify whether the pose error between a synthetic rendering and an observed image of the same object can be corrected by the refiner.
  arXiv  Detail & Related papers  (2022-12-13T19:30:03Z)
• Shape, Pose, and Appearance from a Single Image via Bootstrapped Radiance Field Inversion [54.151979979158085]
  We introduce a principled end-to-end reconstruction framework for natural images, where accurate ground-truth poses are not available. We leverage an unconditional 3D-aware generator, to which we apply a hybrid inversion scheme where a model produces a first guess of the solution. Our framework can de-render an image in as few as 10 steps, enabling its use in practical scenarios.
  arXiv  Detail & Related papers  (2022-11-21T17:42:42Z)
• SPARC: Sparse Render-and-Compare for CAD model alignment in a single RGB image [21.77811443143683]
  Estimating 3D shapes and poses of static objects from a single image has important applications for robotics, augmented reality and digital content creation. We demonstrate that a sparse, iterative, render-and-compare approach is more accurate and robust than relying on normalised object coordinates. Our alignment procedure converges after just 3 iterations, improving the state-of-the-art performance on the challenging real-world dataset ScanNet.
  arXiv  Detail & Related papers  (2022-10-03T16:02:10Z)
• ROCA: Robust CAD Model Retrieval and Alignment from a Single Image [22.03752392397363]
  We present ROCA, a novel end-to-end approach that retrieves and aligns 3D CAD models from a shape database to a single input image. experiments on challenging, real-world imagery from ScanNet show that ROCA significantly improves on state of the art, from 9.5% to 17.6% in retrieval-aware CAD alignment accuracy.
  arXiv  Detail & Related papers  (2021-12-03T16:02:32Z)
• Patch2CAD: Patchwise Embedding Learning for In-the-Wild Shape Retrieval from a Single Image [58.953160501596805]
  We propose a novel approach towards constructing a joint embedding space between 2D images and 3D CAD models in a patch-wise fashion. Our approach is more robust than state of the art in real-world scenarios without any exact CAD matches.
  arXiv  Detail & Related papers  (2021-08-20T20:58:52Z)
• Mask2CAD: 3D Shape Prediction by Learning to Segment and Retrieve [54.054575408582565]
  We propose to leverage existing large-scale datasets of 3D models to understand the underlying 3D structure of objects seen in an image. We present Mask2CAD, which jointly detects objects in real-world images and for each detected object, optimize for the most similar CAD model and its pose. This produces a clean, lightweight representation of the objects in an image.
  arXiv  Detail & Related papers  (2020-07-26T00:08:37Z)

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.