2nd Place Solution for IJCAI-PRICAI 2020 3D AI Challenge: 3D Object Reconstruction from A Single Image

TL;DR

This paper presents a top-performing solution for 3D object reconstruction from a single image, based on an enhanced AtlasNet model, with extensive experimental analysis to optimize performance.

Contribution

The authors develop an improved AtlasNet variant for single-image 3D reconstruction and provide detailed insights into key implementation choices affecting results.

Findings

Achieved 2nd place in IJCAI-PRICAI 2020 3D AI Challenge

Optimized decoder design and normalization settings significantly improve reconstruction quality

Provided comprehensive analysis of different sampling and projection methods

Abstract

In this paper, we present our solution for the {\it IJCAI--PRICAI--20 3D AI Challenge: 3D Object Reconstruction from A Single Image}. We develop a variant of AtlasNet that consumes single 2D images and generates 3D point clouds through 2D to 3D mapping. To push the performance to the limit and present guidance on crucial implementation choices, we conduct extensive experiments to analyze the influence of decoder design and different settings on the normalization, projection, and sampling methods. Our method achieves 2nd place in the final track with a score of $70.88$, a chamfer distance of $36.87$, and a mean f-score of $59.18$. The source code of our method will be available at https://github.com/em-data/Enhanced_AtlasNet_3DReconstruction.