paper-with-me

홈 › Papers

SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration

2023-12-08 · CVPR 2024 1 · Xu Cao, Takafumi Taketomi

We present SuperNormal, a fast, high-fidelity approach to multi-view 3D reconstruction using surface normal maps. With a few minutes, SuperNormal produces detailed surfaces on par with 3D scanners. We harness volume rendering to optimize a neural signed distance function (SDF) powered by multi-resolution hash encoding. To accelerate training, we propose directional finite difference and patch-based ray marching to approximate the SDF gradients numerically. While not compromising reconstruction quality, this strategy is nearly twice as efficient as analytical gradients and about three times faster than axis-aligned finite difference. Experiments on the benchmark dataset demonstrate the superiority of SuperNormal in efficiency and accuracy compared to existing multi-view photometric stereo methods. On our captured objects, SuperNormal produces more fine-grained geometry than recent neural 3D reconstruction methods.

📄 PDF Abstract BibTeX arXiv:2312.04803

Code (1)

CyberAgentAILab/SuperNormal 공식 구현 pytorch

Tasks

3D ReconstructionMulti-View 3D ReconstructionSurface Reconstruction

Similar Papers 제목 키워드 기반

NPLMV-PS: Neural Point-Light Multi-View Photometric Stereo

2024-05-20 · Fotios Logothetis, Ignas Budvytis, Roberto Cipolla

In this work we present a novel multi-view photometric stereo (MVPS) method. Like many works in 3D reconstruction we are leveraging neural shape representations and learnt renderers. However, our work differs from the st…

3D ReconstructionNeRFNeural Rendering

PMNI: Pose-free Multi-view Normal Integration for Reflective and Textureless Surface Reconstruction

2025-04-11 · CVPR 2025 1 · Mingzhi Pei, Xu Cao, Xiangyi Wang, Heng Guo 외

Reflective and textureless surfaces remain a challenge in multi-view 3D reconstruction.Both camera pose calibration and shape reconstruction often fail due to insufficient or unreliable cross-view visual features. To add…

Surface Reconstruction

Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs

2024-05-06 · Jordan Dotzel, Yuzong Chen, Bahaa Kotb, Sushma Prasad 외

The increasing size of large language models (LLMs) traditionally requires low-precision integer formats to meet strict latency and power demands. Yet recently, alternative formats such as Normal Float (NF4) have increas…

Quantization

Multi-view Normal and Distance Guidance Gaussian Splatting for Surface Reconstruction

2025-08-11 · Bo Jia, Yanan Guo, Ying Chang, Benkui Zhang 외 arxiv

3D Gaussian Splatting (3DGS) achieves remarkable results in the field of surface reconstruction. However, when Gaussian normal vectors are aligned within the single-view projection plane, while the geometry appears reaso…

Just Look at the Image: Viewpoint-Specific Surface Normal Prediction for Improved Multi-View Reconstruction

2016-06-01 · CVPR 2016 6 · Silvano Galliani, Konrad Schindler

We present a multi-view reconstruction method that combines conventional multi-view stereo (MVS) with appearance-based normal prediction, to obtain dense and accurate 3D surface models. Reliable surface normals reconstru…