Instance-incremental Scene Graph Generation from Real-world Point Clouds via Normalizing Flows

TL;DR

This paper introduces a novel task of instance-incremental scene graph generation from real-world point clouds, using a normalizing flows-based framework to generate and insert new object instances into 3D scenes, with state-of-the-art results.

Contribution

It proposes a new task and a 3D autoregressive normalizing flows model (3D-ANF) for incremental scene graph generation from point clouds, addressing real-world complexity.

Findings

Achieves state-of-the-art performance on 3DSSG-O27R16 and GPL3D datasets.

Successfully generates reliable novel scene graphs from real-world point clouds.

Demonstrates the effectiveness of normalizing flows in 3D scene graph generation.

Abstract

This work introduces a new task of instance-incremental scene graph generation: Given a scene of the point cloud, representing it as a graph and automatically increasing novel instances. A graph denoting the object layout of the scene is finally generated. It is an important task since it helps to guide the insertion of novel 3D objects into a real-world scene in vision-based applications like augmented reality. It is also challenging because the complexity of the real-world point cloud brings difficulties in learning object layout experiences from the observation data (non-empty rooms with labeled semantics). We model this task as a conditional generation problem and propose a 3D autoregressive framework based on normalizing flows (3D-ANF) to address it. First, we represent the point cloud as a graph by extracting the label semantics and contextual relationships. Next, a model based on normalizing flows is introduced to map the conditional generation of graphic elements into the Gaussian process. The mapping is invertible. Thus, the real-world experiences represented in the observation data can be modeled in the training phase, and novel instances can be autoregressively generated based on the Gaussian process in the testing phase. To evaluate the performance of our method sufficiently, we implement this new task on the indoor benchmark dataset 3DSSG-O27R16 and our newly proposed graphical dataset of outdoor scenes GPL3D. Experiments show that our method generates reliable novel graphs from the real-world point cloud and achieves state-of-the-art performance on the datasets.