SketchBodyNet: A Sketch-Driven Multi-faceted Decoder Network for 3D Human Reconstruction

TL;DR

This paper introduces SketchBodyNet, a novel neural network that reconstructs 3D human shapes from freehand sketches by leveraging a multi-faceted decoder with attention mechanisms, outperforming existing methods.

Contribution

The paper presents a new sketch-driven multi-faceted decoder network with attention modules for 3D human reconstruction from sketches, and introduces a large-scale dataset for training and evaluation.

Findings

SketchBodyNet outperforms existing methods in 3D reconstruction accuracy.

The multi-faceted decoder effectively predicts camera, shape, and pose parameters.

The dataset of 26k sketches enhances training and evaluation capabilities.

Abstract

Reconstructing 3D human shapes from 2D images has received increasing attention recently due to its fundamental support for many high-level 3D applications. Compared with natural images, freehand sketches are much more flexible to depict various shapes, providing a high potential and valuable way for 3D human reconstruction. However, such a task is highly challenging. The sparse abstract characteristics of sketches add severe difficulties, such as arbitrariness, inaccuracy, and lacking image details, to the already badly ill-posed problem of 2D-to-3D reconstruction. Although current methods have achieved great success in reconstructing 3D human bodies from a single-view image, they do not work well on freehand sketches. In this paper, we propose a novel sketch-driven multi-faceted decoder network termed SketchBodyNet to address this task. Specifically, the network consists of a backbone and three separate attention decoder branches, where a multi-head self-attention module is exploited in each decoder to obtain enhanced features, followed by a multi-layer perceptron. The multi-faceted decoders aim to predict the camera, shape, and pose parameters, respectively, which are then associated with the SMPL model to reconstruct the corresponding 3D human mesh. In learning, existing 3D meshes are projected via the camera parameters into 2D synthetic sketches with joints, which are combined with the freehand sketches to optimize the model. To verify our method, we collect a large-scale dataset of about 26k freehand sketches and their corresponding 3D meshes containing various poses of human bodies from 14 different angles. Extensive experimental results demonstrate our SketchBodyNet achieves superior performance in reconstructing 3D human meshes from freehand sketches.