Monocular 3D reconstruction using Chamfer distance twice with CD2

Monocular 3D reconstruction using Chamfer distance twice with CD2

CD2: Fine-grained 3D Mesh Reconstruction With Twice Chamfer Distance
arXiv paper abstract https://arxiv.org/abs/2206.00447
arXiv PDF paper https://arxiv.org/pdf/2206.00447.pdf
GitHub https://github.com/ThibaultGROUEIX/ChamferDistancePytorch

Monocular 3D reconstruction is to reconstruct the shape of object and its other information from a single RGB image.

In 3D reconstruction, polygon mesh … is the most prevalent expression form obtained from deep learning models.

… state-of-the-art schemes fail to directly generate well-structured meshes … identify that most meshes have severe Vertices Clustering (VC) and Illegal Twist (IT) problems.

… the inappropriate usage of Chamfer Distance (CD) loss is a root cause of VC and IT problems in deep learning model.

… propose a fine-grained reconstruction method CD2 by employing Chamfer distance twice to perform a plausible and adaptive deformation.

… CD2 directly generates a well-structured mesh and outperforms others in terms of several quantitative metrics.

Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website

LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b

Photo by Denis's Pixels on Unsplash