Filters








350 Hits in 5.1 sec

Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video [article]

Jia-Wang Bian, Zhichao Li, Naiyan Wang, Huangying Zhan, Chunhua Shen, Ming-Ming Cheng, Ian Reid
2019 arXiv   pre-print
Recent work has shown that CNN-based depth and ego-motion estimators can be learned using unlabelled monocular videos.  ...  Moreover, we show that our ego-motion network is able to predict a globally scale-consistent camera trajectory for long video sequences, and the resulting visual odometry accuracy is competitive with the  ...  Conclusion This paper presents an unsupervised learning framework for scale-consistent depth and ego-motion estimation.  ... 
arXiv:1908.10553v2 fatcat:m6ext2y33bhqfdbuejonuvtyhu

Unsupervised Learning for Depth, Ego-Motion, and Optical Flow Estimation Using Coupled Consistency Conditions

Ji-Hun Mun, Moongu Jeon, Byung-Geun Lee
2019 Sensors  
Herein, we propose an unsupervised learning architecture under coupled consistency conditions to estimate the depth, ego-motion, and optical flow.  ...  First, we compare the optical flows, which are estimated from both the depth plus ego-motion and flow estimation network.  ...  Flow Consistency with Depth and Ego-Motion We build a flow consistency block that estimates a depth from a monocular video.  ... 
doi:10.3390/s19112459 fatcat:j6vfrfdbqrcdlerilxhjdbpkjq

DiPE: Deeper into Photometric Errors for Unsupervised Learning of Depth and Ego-motion from Monocular Videos [article]

Hualie Jiang, Laiyan Ding, Zhenglong Sun, Rui Huang
2020 arXiv   pre-print
Unsupervised learning of depth and ego-motion from unlabelled monocular videos has recently drawn great attention, which avoids the use of expensive ground truth in the supervised one.  ...  The overall system achieves state-of-theart performance on both depth and ego-motion estimation.  ...  CONCLUSION In this paper, we have demonstrated that carefully processing the photometric errors for unsupervised learning of depth and ego-motion from monocular videos can successfully solve the intrinsic  ... 
arXiv:2003.01360v3 fatcat:p5toyxdmhfbh3mnvy6tgsmryfi

Unsupervised Learning of Depth and Ego-Motion from Monocular Video Using 3D Geometric Constraints [article]

Reza Mahjourian, Martin Wicke, Anelia Angelova
2018 arXiv   pre-print
We present a novel approach for unsupervised learning of depth and ego-motion from monocular video.  ...  Unsupervised learning removes the need for separate supervisory signals (depth or ego-motion ground truth, or multi-view video).  ...  videos.  ... 
arXiv:1802.05522v2 fatcat:vq7w353yvbeibbmnmrfhv457mi

Unsupervised Learning of Depth and Ego-Motion from Monocular Video Using 3D Geometric Constraints

Reza Mahjourian, Martin Wicke, Anelia Angelova
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
We present a novel approach for unsupervised learning of depth and ego-motion from monocular video.  ...  Unsupervised learning removes the need for separate supervisory signals (depth or ego-motion ground truth, or multi-view video).  ...  videos.  ... 
doi:10.1109/cvpr.2018.00594 dblp:conf/cvpr/MahjourianWA18 fatcat:vvjhfpn4qvdkffujq747pxfk4e

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

Yang Wang, Zhenheng Yang, Peng Wang, Yi Yang, Chenxu Luo, Wei Xu
2018 arXiv   pre-print
Specifically, given two consecutive stereo image pairs from a video, we first estimate depth, camera ego-motion and optical flow from three neural networks.  ...  Then the whole scene is decomposed into moving foreground and static background by compar- ing the estimated optical flow and rigid flow derived from the depth and ego-motion.  ...  Related Work Unsupervised Learning of Depth and Ego-motion The unsupervised learning of depth and ego-motion through monocular videos using deep learning was first achieved in (Zhou et al. 2017b) .  ... 
arXiv:1810.03654v1 fatcat:kejv4uquazao3ilf4ijetbfzjq

Instance-wise Depth and Motion Learning from Monocular Videos [article]

Seokju Lee, Sunghoon Im, Stephen Lin, In So Kweon
2020 arXiv   pre-print
We present an end-to-end joint training framework that explicitly models 6-DoF motion of multiple dynamic objects, ego-motion and depth in a monocular camera setup without supervision.  ...  First, we propose a differentiable forward rigid projection module that plays a key role in our instance-wise depth and motion learning.  ...  Related Works Unsupervised depth and ego-motion learning Several works [48, 14, 41, 28, 35] have studied the inference of depth and ego-motion from monocular sequences. Zhou et al.  ... 
arXiv:1912.09351v2 fatcat:knjqipriwfbldescp36mtp3buq

Masked GANs for Unsupervised Depth and Pose Prediction with Scale Consistency [article]

Chaoqiang Zhao, Gary G. Yen, Qiyu Sun, Chongzhen Zhang, Yang Tang
2021 arXiv   pre-print
This paper proposes a masked generative adversarial network (GAN) for unsupervised monocular depth and ego-motion estimation.The MaskNet and Boolean mask scheme are designed in this framework to eliminate  ...  Previous work has shown that adversarial learning can be used for unsupervised monocular depth and visual odometry (VO) estimation, in which the adversarial loss and the geometric image reconstruction  ...  Learning from monocular videos.  ... 
arXiv:2004.04345v2 fatcat:rc5e2x4rprcwnl2otqrfo6siuu

Unsupervised Video Depth Estimation Based on Ego-motion and Disparity Consensus [article]

Lingtao Zhou, Jiaojiao Fang, Guizhong Liu
2019 arXiv   pre-print
In this paper, we propose a novel unsupervised monocular video depth estimation method in natural scenes by taking advantage of the state-of-the-art method of Zhou et al. which jointly estimates depth  ...  and camera motion.  ...  [14] propose a method jointly learning of ego-motion and depth from unlabeled videos by unsupervised manner with the static scene assumption.  ... 
arXiv:1909.01028v1 fatcat:3sgsalnd6jaaln7wp2iemixawy

Unsupervised Learning of Monocular Depth and Ego-Motion Using Multiple Masks

Guangming Wang, Hesheng Wang, Yiling Liu, Weidong Chen
2019 2019 International Conference on Robotics and Automation (ICRA)  
A new unsupervised learning method of depth and ego-motion using multiple masks from monocular video is proposed in this paper.  ...  The depth estimation network and the ego-motion estimation network are trained according to the constraints of depth and ego-motion without truth values.  ...  [12] conducted an unsupervised joint estimation of depth and ego-motion from monocular video and obtained competitive depth and pose Unsupervised Learning of Monocular Depth and Ego-Motion Using Multiple  ... 
doi:10.1109/icra.2019.8793622 dblp:conf/icra/WangWL019 fatcat:nhfrj6jsbvfpndowbyoqeinni4

Depth Prediction without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videos

Vincent Casser, Soeren Pirk, Reza Mahjourian, Anelia Angelova
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In this work we address unsupervised learning of scene depth and robot ego-motion where supervision is provided by monocular videos, as cameras are the cheapest, least restrictive and most ubiquitous sensor  ...  The main idea is to introduce geometric structure in the learning process, by modeling the scene and the individual objects; camera ego-motion and object motions are learned from monocular videos as input  ...  Main Method The main learning setup is unsupervised learning of depth and ego-motion from monocular video , where the only source of supervision is obtained from the video itself.  ... 
doi:10.1609/aaai.v33i01.33018001 fatcat:4p5kkn52bndubnbrshiz2le62i

Unsupervised Monocular Depth Perception: Focusing on Moving Objects [article]

Hualie Jiang, Laiyan Ding, Zhenglong Sun, Rui Huang
2021 arXiv   pre-print
As a flexible passive 3D sensing means, unsupervised learning of depth from monocular videos is becoming an important research topic.  ...  Extensive experiments on the KITTI dataset and additional experiments on the Cityscapes dataset have verified the proposed approach's effectiveness on depth or ego-motion estimation.  ...  As video sequences are pervasive on vehicles with mounted cameras [15] , more recent unsupervised learning of depth and ego-motion from monocular videos [16] [17] [18] [19] [20] [21] [22] [23] [24] [  ... 
arXiv:2108.13062v1 fatcat:b535zpzfqzbmbcyyrm5ykqhe6y

Depth Prediction Without the Sensors: Leveraging Structure for Unsupervised Learning from Monocular Videos [article]

Vincent Casser, Soeren Pirk, Reza Mahjourian, Anelia Angelova
2018 arXiv   pre-print
In this work we address unsupervised learning of scene depth and robot ego-motion where supervision is provided by monocular videos, as cameras are the cheapest, least restrictive and most ubiquitous sensor  ...  The main idea is to introduce geometric structure in the learning process, by modeling the scene and the individual objects; camera ego-motion and object motions are learned from monocular videos as input  ...  Main Method The main learning setup is unsupervised learning of depth and ego-motion from monocular video , where the only source of supervision is obtained from the video itself.  ... 
arXiv:1811.06152v1 fatcat:zkrl6iv4wbbrroepf5tkpfynzu

Semantics-Driven Unsupervised Learning for Monocular Depth and Ego-Motion Estimation [article]

Xiaobin Wei, Jianjiang Feng, Jie Zhou
2020 arXiv   pre-print
We propose a semantics-driven unsupervised learning approach for monocular depth and ego-motion estimation from videos in this paper.  ...  Experimental results on the KITTI dataset show that our method achieves good performance in both depth and ego-motion estimation tasks.  ...  GeoNet [51] is a jointly unsupervised learning framework for monocular depth, optical flow and ego-motion estimation from videos, which uses separate components to learn the rigid flow and object motion  ... 
arXiv:2006.04371v1 fatcat:v7wrccts4zh3pbsggqfn552ouq

Unsupervised Learning of Depth from Monocular Videos Using 3D-2D Corresponding Constraints

Fusheng Jin, Yu Zhao, Chuanbing Wan, Ye Yuan, Shuliang Wang
2021 Remote Sensing  
This paper proposes a depth prediction method for AMP based on unsupervised learning, which can learn from video sequences and simultaneously estimate the depth structure of the scene and the ego-motion  ...  Compared with the existing unsupervised learning methods, our method makes the spatial correspondence among pixel points consistent with the image area by smoothing the 3D corresponding vector field based  ...  to the target using the predicted depth and ego-motion, which make it possible to train the network using monocular video sequence.  ... 
doi:10.3390/rs13091764 fatcat:hin47ftalfghfo5ixenjh7rnei
« Previous Showing results 1 — 15 out of 350 results