mcc3d.github.io - Multiview Compressive Coding for 3D Reconstruction

Description: MCC.

Example domain paragraphs

MCC is a new approach for 3D reconstruction from a single RGB-D image.

A central goal of visual recognition is to understand objects and scenes from a single image. 2D recognition has witnessed tremendous progress thanks to large-scale learning and general-purpose representations. But, 3D poses new challenges stemming from occlusions not depicted in the image. Prior works try to overcome these by inferring from multiple views or rely on scarce CAD models and category-specific priors which hinder scaling to novel settings. In this work, we explore single-view 3D reconstruction

MCC adopts a simple encoder-decoder architecture. The input RGB-D image is fed to the encoder to produce an input encoding. The decoder inputs a query 3D point, along with the input encoding, to predict its occupancy probability and RGB color. This is illustrated in the figure below. MCC requires only points for supervision. This means that we can learn from large RGB-D datasets. The input channel D is read from depth sensors, as in iPhones, or computed by off-the-shelf depth models, e.g. MiDas , or compute

Links to mcc3d.github.io (5)

gkioxari.github.io Georgia Gkioxari
feichtenhofer.github.io Christoph Feichtenhofer
chaoyuan.org Chao-Yuan Wu
chaoyuaw.github.io Chao-Yuan Wu
numcc.github.io NU-MCC: Multiview Compressive Coding with Neighborhood Decoder and Repulsive UDF