Dense Depth from Event Focal Stack
Kenta Horikawa, Mariko Isogawa, Hideo Saito, Shohei Mori

TL;DR
This paper introduces a neural network-based approach for dense depth estimation from event streams generated by a focal stack, trained on synthetic data to handle diverse 3D scenes and real-world domain gaps.
Contribution
It presents a novel method using synthesized event focal stacks for training, enabling accurate depth estimation from event cameras in various scenes.
Findings
Outperforms depth-from-defocus methods on synthetic and real datasets
Effective training on synthetic data reduces domain gap issues
Demonstrates superior depth estimation accuracy
Abstract
We propose a method for dense depth estimation from an event stream generated when sweeping the focal plane of the driving lens attached to an event camera. In this method, a depth map is inferred from an ``event focal stack'' composed of the event stream using a convolutional neural network trained with synthesized event focal stacks. The synthesized event stream is created from a focal stack generated by Blender for any arbitrary 3D scene. This allows for training on scenes with diverse structures. Additionally, we explored methods to eliminate the domain gap between real event streams and synthetic event streams. Our method demonstrates superior performance over a depth-from-defocus method in the image domain on synthetic and real datasets.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsScientific Computing and Data Management · Explainable Artificial Intelligence (XAI)
MethodsRoIAlign · Softmax · RoIPool
