Language Driven Occupancy Prediction

TL;DR

LOcc introduces a semantic transitive labeling pipeline for open-vocabulary 3D occupancy prediction, leveraging image-to-text transfer to improve accuracy and reduce manual labeling, and demonstrates superior performance over existing methods.

Contribution

The paper presents a novel semantic transitive labeling pipeline and a new LOcc framework that enhances open-vocabulary 3D occupancy prediction with dense language supervision.

Findings

LOcc outperforms state-of-the-art zero-shot methods on Occ3D-nuScenes.

Semantic transitive labeling improves pseudo-label accuracy.

LOcc reduces reliance on manual annotations.

Abstract

We introduce LOcc, an effective and generalizable framework for open-vocabulary occupancy (OVO) prediction. Previous approaches typically supervise the networks through coarse voxel-to-text correspondences via image features as intermediates or noisy and sparse correspondences from voxel-based model-view projections. To alleviate the inaccurate supervision, we propose a semantic transitive labeling pipeline to generate dense and fine-grained 3D language occupancy ground truth. Our pipeline presents a feasible way to dig into the valuable semantic information of images, transferring text labels from images to LiDAR point clouds and ultimately to voxels, to establish precise voxel-to-text correspondences. By replacing the original prediction head of supervised occupancy models with a geometry head for binary occupancy states and a language head for language features, LOcc effectively uses the generated language ground truth to guide the learning of 3D language volume. Through extensive experiments, we demonstrate that our transitive semantic labeling pipeline can produce more accurate pseudo-labeled ground truth, diminishing labor-intensive human annotations. Additionally, we validate LOcc across various architectures, where all models consistently outperform state-of-the-art zero-shot occupancy prediction approaches on the Occ3D-nuScenes dataset.