Fast Kernel Scene Flow
This work addresses the need for fast and generalizable scene flow estimation in robotics and autonomous driving, though it is incremental as it builds on classical kernel representations rather than deep learning.
The paper tackles the problem of efficient scene flow estimation from dense lidar point clouds by introducing a kernel-based method that achieves near real-time performance (~150-170 ms) with competitive accuracy on large-scale datasets.
In contrast to current state-of-the-art methods, such as NSFP [25], which employ deep implicit neural functions for modeling scene flow, we present a novel approach that utilizes classical kernel representations. This representation enables our approach to effectively handle dense lidar points while demonstrating exceptional computational efficiency -- compared to recent deep approaches -- achieved through the solution of a linear system. As a runtime optimization-based method, our model exhibits impressive generalizability across various out-of-distribution scenarios, achieving competitive performance on large-scale lidar datasets. We propose a new positional encoding-based kernel that demonstrates state-of-the-art performance in efficient lidar scene flow estimation on large-scale point clouds. An important highlight of our method is its near real-time performance (~150-170 ms) with dense lidar data (~8k-144k points), enabling a variety of practical applications in robotics and autonomous driving scenarios.