2nd Place Solution for IJCAI-PRICAI 2020 3D AI Challenge: 3D Object Reconstruction from A Single Image
This work addresses 3D reconstruction for computer vision applications, but it is incremental as it builds on an existing method with optimizations.
The paper tackles 3D object reconstruction from a single image by developing a variant of AtlasNet, achieving 2nd place in a challenge with scores of 70.88, chamfer distance of 36.87, and mean f-score of 59.18.
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.