Learning End-To-End Scene Flow by Distilling Single Tasks Knowledge

11/22/2019
by   Filippo Aleotti, et al.
27

Scene flow is a challenging task aimed at jointly estimating the 3D structure and motion of the sensed environment. Although deep learning solutions achieve outstanding performance in terms of accuracy, these approaches divide the whole problem into standalone tasks (stereo and optical flow) addressing them with independent networks. Such a strategy dramatically increases the complexity of the training procedure and requires power-hungry GPUs to infer scene flow barely at 1 FPS. Conversely, we propose DWARF, a novel and lightweight architecture able to infer full scene flow jointly reasoning about depth and optical flow easily and elegantly trainable end-to-end from scratch. Moreover, since ground truth images for full scene flow are scarce, we propose to leverage on the knowledge learned by networks specialized in stereo or flow, for which much more data are available, to distill proxy annotations. Exhaustive experiments show that i) DWARF runs at about 10 FPS on a single high-end GPU and about 1 FPS on NVIDIA Jetson TX2 embedded at KITTI resolution, with moderate drop in accuracy compared to 10x deeper models, ii) learning from many distilled samples is more effective than from the few, annotated ones available. Code available at: https://github.com/FilippoAleotti/Dwarf-Tensorflow

READ FULL TEXT

page 1

page 3

page 6

page 9

page 10

page 11

page 12

page 13

research
10/08/2018

Joint Unsupervised Learning of Optical Flow and Depth by Watching Stereo Videos

Learning depth and optical flow via deep neural networks by watching vid...
research
11/16/2020

EffiScene: Efficient Per-Pixel Rigidity Inference for Unsupervised Joint Learning of Optical Flow, Depth, Camera Pose and Motion Segmentation

This paper addresses the challenging unsupervised scene flow estimation ...
research
03/03/2023

Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereo

While recent methods for motion and stereo estimation recover an unprece...
research
10/01/2019

Real-Time Semantic Stereo Matching

Scene understanding is paramount in robotics, self-navigation, augmented...
research
04/12/2019

PWOC-3D: Deep Occlusion-Aware End-to-End Scene Flow Estimation

In the last few years, convolutional neural networks (CNNs) have demonst...
research
03/21/2023

Learning Optical Flow and Scene Flow with Bidirectional Camera-LiDAR Fusion

In this paper, we study the problem of jointly estimating the optical fl...
research
05/22/2019

Bridging Stereo Matching and Optical Flow via Spatiotemporal Correspondence

Stereo matching and flow estimation are two essential tasks for scene un...

Please sign up or login with your details

Forgot password? Click here to reset