Unsupervised Depth Completion from Visual Inertial Odometry (RAL 2020 and ICRA 2020)

preview_player
Показать описание
Unsupervised Depth Completion from Visual Inertial Odometry

Authors: Alex Wong, Xiaohan Fei, Stephanie Tsuei, and Stefano Soatto

Published in the Robotics and Automation Letters (RA-L) 2020 and the
proceedings of International Conference on Robotics and Automation (ICRA) 2020.

In this talk, we present an unsupervised (no need for human supervision) method for learning to recover 3D scene geometry (depth) from images captured by cameras, and sparse point clouds produced by lidar or tracked by visual inertial odometry systems.

Checkout our follow up work on this:
Learning Topology from Synthetic Data for Unsupervised Depth Completion (RAL 2021 and ICRA 2021)
  1. Alex Wong
  2. computer vision
  3. machine learning
  4. unsupervised learning
  5. 3d vision
  6. sensor fusion
Рекомендации по теме
Unsupervised Depth Completion 0:13:25

Unsupervised Depth Completion from Visual Inertial Odometry (RAL 2020 and ICRA 2020)

Struct-MDC: Mesh-Refined Unsupervised 0:01:51

Struct-MDC: Mesh-Refined Unsupervised Depth Completion Leveraging Structural Regularities

Learning Topology from 0:17:51

Learning Topology from Synthetic Data for Unsupervised Depth Completion (RAL 2021 and ICRA 2021)

Unsupervised Monocular Depth 0:04:51

Unsupervised Monocular Depth Estimation with Left-Right Consistency

Teaser: Monocular Visual-Inertial 0:02:20

Teaser: Monocular Visual-Inertial Depth Estimation (ICRA'23)

Robust Monocular Visual-Inertial 0:12:18

Robust Monocular Visual-Inertial Depth Completion for Embedded Systems

NeurIPS2019 depth completion 0:00:45

NeurIPS2019 depth completion video

Self-supervised Sparse-to-Dense: Self-supervised 0:02:18

Self-supervised Sparse-to-Dense: Self-supervised Depth Completion from LiDAR and Monocular Camera

Unsupervised Monocular Depth 0:13:34

Unsupervised Monocular Depth Estimation CNN Robust to Training Data Diversity

Robust Monocular Visual-Inertial 0:01:15

Robust Monocular Visual-Inertial Depth Completion for Embedded Systems

Depth Completion Auto-Encoder 0:08:35

Depth Completion Auto-Encoder

Uncertainty-Aware CNNs for 0:01:00

Uncertainty-Aware CNNs for Depth Completion: Uncertainty from Beginning to End

Self-Supervised Depth Completion 0:03:58

Self-Supervised Depth Completion for Active Stereo [ICRA presentation]

From Depth What 0:01:00

From Depth What Can You See? Depth Completion via Auxiliary Image Reconstruction

Balanced Depth Completion 0:02:18

Balanced Depth Completion between Dense Depth Inference and Sparse Range Measurements (IROS 2020)

Unsupervised Monocular Depth 0:12:44

Unsupervised Monocular Depth Estimation With Left-Right Consistency

3D Depth From 0:02:07

3D Depth From a Single Photograph | Two Minute Papers #54

AugUndo: Scaling Up 0:07:24

AugUndo: Scaling Up Augmentations for Monocular Depth Completion and Estimation

Elimination of Incorrect 0:11:38

Elimination of Incorrect Depth Points for Depth Completion

[ICCV 2021] Domain 0:05:19

[ICCV 2021] Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation

The Edge of 0:01:00

The Edge of Depth: Explicit Constraints Between Segmentation and Depth

SLAMcore Labs - 0:00:18

SLAMcore Labs - AI enabled depth completion

Real-time Wide-baseline Place 0:03:04

Real-time Wide-baseline Place Recognition using Depth Completion - RA-L19

Alex Z. Zhu: 0:21:37

Alex Z. Zhu: Unsupervised Learning of Optical Flow and Camera Motion from Event Data