Reconstructing vehicles from orthographic drawings using deep neural networks

TL;DR

This paper presents deep neural network methods for reconstructing vehicles from orthographic drawings, introducing novel algorithms, a new dataset, and demonstrating high-quality, generalizable 3D reconstructions.

Contribution

It introduces two algorithms for extracting multiple views from a single image and a new dataset for vehicle reconstruction, advancing the state-of-the-art in object reconstruction from orthographic drawings.

Findings

The proposed system produces detailed and plausible vehicle reconstructions.

The neural network generalizes well to real-world inputs.

The approach outperforms existing methods in quality and accuracy.

Abstract

This paper explores the current state-of-the-art of object reconstruction from multiple orthographic drawings using deep neural networks. It proposes two algorithms to extract multiple views from a single image. The paper proposes a system based on pixel-aligned implicit functions (PIFu) and develops an advanced sampling strategy to generate signed distance samples. It also compares this approach to depth map regression from multiple views. Additionally, the paper uses a novel dataset for vehicle reconstruction from the racing game Assetto Corsa, which features higher quality models than the commonly used ShapeNET dataset. The trained neural network generalizes well to real-world inputs and creates plausible and detailed reconstructions.