ForecastOcc: Vision-based Semantic Occupancy Forecasting

TL;DR

ForecastOcc is a pioneering vision-based framework that jointly predicts future semantic occupancy states directly from camera images, enhancing scene understanding for autonomous driving without relying on external maps.

Contribution

It introduces the first joint semantic occupancy forecasting model from images, with a novel architecture including cross-attention and view transformer modules, and establishes new benchmarks.

Findings

Outperforms baseline methods on Occ3D-nuScenes and SemanticKITTI datasets.

Provides the first benchmarks for monocular semantic occupancy forecasting.

Demonstrates accurate, future-aware scene predictions with rich semantic information.

Abstract

Autonomous driving requires forecasting both geometry and semantics over time to effectively reason about future environment states. Existing vision-based occupancy forecasting methods focus on motion-related categories such as static and dynamic objects, while semantic information remains largely absent. Recent semantic occupancy forecasting approaches address this gap but rely on past occupancy predictions obtained from separate networks. This makes current methods sensitive to error accumulation and prevents learning spatio-temporal features directly from images. In this work, we present ForecastOcc, the first framework for vision-based semantic occupancy forecasting that jointly predicts future occupancy states and semantic categories. Our framework yields semantic occupancy forecasts for multiple horizons directly from past camera images, without relying on externally estimated maps. We evaluate ForecastOcc in two complementary settings: multi-view forecasting on the Occ3D-nuScenes dataset and monocular forecasting on SemanticKITTI, where we establish the first benchmark for this task. We introduce the first baselines by adapting two 2D forecasting modules within our framework. Importantly, we propose a novel architecture that incorporates a temporal cross-attention forecasting module, a 2D-to-3D view transformer, a 3D encoder for occupancy prediction, and a semantic occupancy head for voxel-level forecasts across multiple horizons. Extensive experiments on both datasets show that ForecastOcc consistently outperforms baselines, yielding semantically rich, future-aware predictions that capture scene dynamics and semantics critical for autonomous driving.