paper-with-me

Papers

Robust 3D Shape Reconstruction in Zero-Shot from a Single Image in the Wild

2024-03-21 · CVPR 2025 1 · Junhyeong Cho, Kim Youwang, Hunmin Yang, Tae-Hyun Oh

Recent monocular 3D shape reconstruction methods have shown promising zero-shot results on object-segmented images without any occlusions. However, their effectiveness is significantly compromised in real-world conditions, due to imperfect object segmentation by off-the-shelf models and the prevalence of occlusions. To effectively address these issues, we propose a unified regression model that integrates segmentation and reconstruction, specifically designed for occlusion-aware 3D shape reconstruction. To facilitate its reconstruction in the wild, we also introduce a scalable data synthesis pipeline that simulates a wide range of variations in objects, occluders, and backgrounds. Training on our synthetic data enables the proposed model to achieve state-of-the-art zero-shot results on real-world images, using significantly fewer parameters than competing approaches.

📄 PDF Abstract BibTeX arXiv:2403.14539

Code (0)

등록된 구현이 없습니다.

Tasks

3D Shape ReconstructionObjectSegmentationSemantic Segmentation

Similar Papers 제목 키워드 기반

ZeroShape: Regression-based Zero-shot Shape Reconstruction

2023-12-21 · CVPR 2024 1 · Zixuan Huang, Stefan Stojanov, Anh Thai, Varun Jampani 외

We study the problem of single-image zero-shot 3D shape reconstruction. Recent works learn zero-shot shape reconstruction through generative modeling of 3D assets, but these models are computationally expensive at train …

3D Shape ReconstructionComputational Efficiencyregression

Shape Non-rigid Kinematics (SNK): A Zero-Shot Method for Non-Rigid Shape Matching via Unsupervised Functional Map Regularized Reconstruction

2024-03-11 · NeurIPS 2023 11 · Souhaib Attaiki, Maks Ovsjanikov

We present Shape Non-rigid Kinematics (SNK), a novel zero-shot method for non-rigid shape matching that eliminates the need for extensive training or ground truth data. SNK operates on a single pair of shapes, and employ…

Decoder

Single-shot 3D shape reconstruction using deep convolutional neural networks

2019-09-17 · Hieu Nguyen, Hui Li, Qiang Qiu, Yuzeng Wang 외

A robust single-shot 3D shape reconstruction technique integrating the fringe projection profilometry (FPP) technique with the deep convolutional neural networks (CNNs) is proposed in this letter. The input of the propos…

3D Shape Reconstruction

Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors

2022-09-30 · Zhen Xing, Hengduo Li, Zuxuan Wu, Yu-Gang Jiang

The performance of existing single-view 3D reconstruction methods heavily relies on large-scale 3D annotations. However, such annotations are tedious and expensive to collect. Semi-supervised learning serves as an altern…

3D Reconstructionimage-classificationImage ClassificationObject Reconstruction+2

Learning to Recover 3D Scene Shape from a Single Image

2020-12-17 · CVPR 2021 1 · Wei Yin, Jianming Zhang, Oliver Wang, Simon Niklaus 외

Despite significant progress in monocular depth estimation in the wild, recent state-of-the-art methods cannot be used to recover accurate 3D scene shape due to an unknown depth shift induced by shift-invariant reconstru…

3D Scene ReconstructionDepth EstimationDepth PredictionIndoor Monocular Depth Estimation+2