Anytime Stereo Image Depth Estimation on Mobile Devices
arXiv:1810.11408
Abstract
Many applications of stereo depth estimation in robotics require the generation of accurate disparity maps in real time under significant computational constraints. Current state-of-the-art algorithms force a choice between either generating accurate mappings at a slow pace, or quickly generating inaccurate ones, and additionally these methods typically require far too many parameters to be usable on power- or memory-constrained devices. Motivated by these shortcomings, we propose a novel approach for disparity prediction in the anytime setting. In contrast to prior work, our end-to-end learned approach can trade off computation and accuracy at inference time. Depth estimation is performed in stages, during which the model can be queried at any time to output its current best estimate. Our final model can process 1242375 resolution images within a range of 10-35 FPS on an NVIDIA Jetson TX2 module with only marginal increases in error -- using two orders of magnitude fewer parameters than the most competitive baseline. The source code is available at https://github.com/mileyan/AnyNet .
arXiv commentAccepted by ICRA2019
References in corpus (11)
- Adam: A Method for Stochastic Optimization
- FlowNet: Learning Optical Flow with Convolutional Networks
- End-to-End Learning of Geometry and Context for Deep Stereo Regression
- Convolutional Neural Fabrics
- Cost-Sensitive Tree of Classifiers
- Deciding How to Decide: Dynamic Routing in Artificial Neural Networks
- The Greedy Miser: Learning under Test-time Budgets
- BlockDrop: Dynamic Inference Paths in Residual Networks
- Learning for Disparity Estimation through Feature Constancy
- Efficient Learning by Directed Acyclic Graph For Resource Constrained Prediction
- StereoNet: Guided Hierarchical Refinement for Real-Time Edge-Aware Depth Prediction