Learning Stixel-based Instance Segmentation
This work addresses instance segmentation for autonomous driving systems by enabling efficient processing of Stixel data, though it is incremental as it adapts existing point cloud techniques to a new domain.
The paper tackles the problem of performing instance segmentation on Stixels, a sparse representation used in autonomous driving, by proposing StixelPointNet, a method that treats Stixels as point clouds and uses a PointNet model to achieve state-of-the-art performance on Stixel-level and faster speeds than pixel-based methods.
Stixels have been successfully applied to a wide range of vision tasks in autonomous driving, recently including instance segmentation. However, due to their sparse occurrence in the image, until now Stixels seldomly served as input for Deep Learning algorithms, restricting their utility for such approaches. In this work we present StixelPointNet, a novel method to perform fast instance segmentation directly on Stixels. By regarding the Stixel representation as unstructured data similar to point clouds, architectures like PointNet are able to learn features from Stixels. We use a bounding box detector to propose candidate instances, for which the relevant Stixels are extracted from the input image. On these Stixels, a PointNet models learns binary segmentations, which we then unify throughout the whole image in a final selection step. StixelPointNet achieves state-of-the-art performance on Stixel-level, is considerably faster than pixel-based segmentation methods, and shows that with our approach the Stixel domain can be introduced to many new 3D Deep Learning tasks.