BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images

TL;DR

BlockGAN is a novel generative model that learns 3D scene representations from unlabelled images, enabling realistic rendering and control over object properties without requiring 3D annotations or multiple views.

Contribution

It introduces a 3D-aware generative framework inspired by graphics pipelines that learns object and scene representations directly from 2D images, capturing occlusion and interaction.

Findings

Learns disentangled object and scene representations.

Generates realistic images with controllable object pose and identity.

Operates without 3D supervision or multiple views.

Abstract

We present BlockGAN, an image generative model that learns object-aware 3D scene representations directly from unlabelled 2D images. Current work on scene representation learning either ignores scene background or treats the whole scene as one object. Meanwhile, work that considers scene compositionality treats scene objects only as image patches or 2D layers with alpha maps. Inspired by the computer graphics pipeline, we design BlockGAN to learn to first generate 3D features of background and foreground objects, then combine them into 3D features for the wholes cene, and finally render them into realistic images. This allows BlockGAN to reason over occlusion and interaction between objects' appearance, such as shadow and lighting, and provides control over each object's 3D pose and identity, while maintaining image realism. BlockGAN is trained end-to-end, using only unlabelled single images, without the need for 3D geometry, pose labels, object masks, or multiple views of the same scene. Our experiments show that using explicit 3D features to represent objects allows BlockGAN to learn disentangled representations both in terms of objects (foreground and background) and their properties (pose and identity).