Diffusion-FS: Multimodal Free-Space Prediction via Diffusion for Autonomous Driving
This addresses the problem of safe navigation in autonomous driving by predicting navigable corridors from camera images, though it is incremental as it builds on existing free-space and diffusion methods.
The paper tackles drivable free-space corridor prediction for autonomous driving by framing it as a monocular image perception task, introducing a self-supervised approach for data generation and a diffusion-based architecture called ContourDiff that denoises contour points, achieving accurate multimodal corridor predictions on nuScenes and CARLA datasets.
Drivable Free-space prediction is a fundamental and crucial problem in autonomous driving. Recent works have addressed the problem by representing the entire non-obstacle road regions as the free-space. In contrast our aim is to estimate the driving corridors that are a navigable subset of the entire road region. Unfortunately, existing corridor estimation methods directly assume a BEV-centric representation, which is hard to obtain. In contrast, we frame drivable free-space corridor prediction as a pure image perception task, using only monocular camera input. However such a formulation poses several challenges as one doesn't have the corresponding data for such free-space corridor segments in the image. Consequently, we develop a novel self-supervised approach for free-space sample generation by leveraging future ego trajectories and front-view camera images, making the process of visual corridor estimation dependent on the ego trajectory. We then employ a diffusion process to model the distribution of such segments in the image. However, the existing binary mask-based representation for a segment poses many limitations. Therefore, we introduce ContourDiff, a specialized diffusion-based architecture that denoises over contour points rather than relying on binary mask representations, enabling structured and interpretable free-space predictions. We evaluate our approach qualitatively and quantitatively on both nuScenes and CARLA, demonstrating its effectiveness in accurately predicting safe multimodal navigable corridors in the image.