Filters








833 Hits in 4.1 sec

Unsupervised Monocular Depth Estimation in Highly Complex Environments [article]

Chaoqiang Zhao, Yang Tang, Qiyu Sun
2021 arXiv   pre-print
In this paper, we investigate the problem of unsupervised monocular depth estimation in certain highly complex scenarios.  ...  Previous unsupervised monocular depth estimation methods mainly focus on the day-time scenario, and their frameworks are driven by warped photometric consistency.  ...  The framework of ITDFA for unsupervised monocular depth estimation in highly complex environments.  ... 
arXiv:2107.13137v1 fatcat:sxupguuzirdxplzofrc7tljkr4

Autonomous quadrotor obstacle avoidance based on dueling double deep recurrent Q-learning with monocular vision [article]

Jiajun Ou, Xiao Guo, Ming Zhu, Wenjie Lou
2020 arXiv   pre-print
In this paper, a novel framework is demonstrated to control a quadrotor flying through crowded environments autonomously with monocular vision.  ...  The sensing module is based on an unsupervised deep learning method.  ...  This framework utilizes an unsupervised deep learning method to estimate depth from the raw RGB monocular image.  ... 
arXiv:2002.03510v2 fatcat:otqli45cxrevvm3nsuvm2lfcli

Real-time 3D Perception of Scene with Monocular Camera

Shadi Saleh, Shanmugapriyan Manoharan, Wolfram Hardt
2020 Embedded Selforganising Systems  
The main objective of depth estimation is to extract a representation of the spatial structure of the environment and to restore the 3D shape and visual appearance of objects in imagery.  ...  in the single image.  ...  Self-supervise Depth Estimation: In this approach, the depth is estimated from the monocular camera based on the stereo information.  ... 
doi:10.14464/ess.v7i2.436 fatcat:ndvpqbr3s5axbhajzwlk3r4dae

Unsupervised Learning-based Depth Estimation aided Visual SLAM Approach [article]

Mingyang Geng, Suning Shang, Bo Ding, Huaimin Wang, Pengfei Zhang, Lei Zhang
2019 arXiv   pre-print
and add training constraints for the task of monocular depth and camera motion estimation.  ...  In this paper, we first present an unsupervised learning framework, which not only uses image reconstruction for supervising but also exploits the pose estimation method to enhance the supervised signal  ...  is performing a complex task and confronted with an unknown environment.  ... 
arXiv:1901.07288v1 fatcat:amnkyh7k3ndjhmnspkfann5msm

Semi-Supervised Learning with Mutual Distillation for Monocular Depth Estimation [article]

Jongbeom Baek, Gyeongnyeon Kim, Seungryong Kim
2022 arXiv   pre-print
We propose a semi-supervised learning framework for monocular depth estimation.  ...  Compared to existing semi-supervised learning methods, which inherit limitations of both sparse supervised and unsupervised loss functions, we achieve the complementary advantages of both loss functions  ...  To learn monocular depth estimation networks in a semisupervised manner, some methods [15] , [16] directly combine the sparse supervised and unsupervised loss functions, but such straightforward approach  ... 
arXiv:2203.09737v1 fatcat:yiobudoacbdtheq6mymnmnkfky

Deep Direct Visual Odometry [article]

Chaoqiang Zhao, Yang Tang, Qiyu Sun, Athanasios V. Vasilakos
2021 arXiv   pre-print
in the unsupervised manner.  ...  Traditional monocular direct visual odometry (DVO) is one of the most famous methods to estimate the ego-motion of robots and map environments from images simultaneously.  ...  [28] propose a novel unsupervised monocular depth estimation network trained on stereo videos, and both the depth, pose, and uncertainty estimated by networks are introduced into DSO [7] , thereby  ... 
arXiv:1912.05101v3 fatcat:ooa4y7aoxbhf3a2mpeanw7iwda

Towards real-time unsupervised monocular depth estimation on CPU [article]

Matteo Poggi, Filippo Aleotti, Fabio Tosi, Stefano Mattoccia
2018 arXiv   pre-print
Similarly to state-of-the-art, we train our network in an unsupervised manner casting depth estimation as an image reconstruction problem.  ...  To the best of our knowledge, it is the first method enabling such performance on CPUs paving the way for effective deployment of unsupervised monocular depth estimation even on embedded systems.  ...  Moreover, despite its reduced complexity it enables unsupervised training and outperforms almost all methodologies proposed in literature for monocular depth estimation including supervised ones.  ... 
arXiv:1806.11430v3 fatcat:ycjwqa7p5vayvk3n2zrfags7by

Unsupervised Monocular Depth Learning in Dynamic Scenes [article]

Hanhan Li, Ariel Gordon, Hang Zhao, Vincent Casser, Anelia Angelova
2020 arXiv   pre-print
We present a method for jointly training the estimation of depth, ego-motion, and a dense 3D translation field of objects relative to the scene, with monocular photometric consistency being the sole source  ...  We show that this regularization alone is sufficient to train monocular depth prediction models that exceed the accuracy achieved in prior work for dynamic scenes, including methods that require semantic  ...  Conclusions This paper presents a novel unsupervised method for depth learning in highly dynamic scenes, which jointly solves for 3D motion maps and depth maps.  ... 
arXiv:2010.16404v2 fatcat:dwf7hypltnbijn3somdt6hv2zu

DeFeat-Net: General Monocular Depth via Simultaneous Unsupervised Representation Learning [article]

Jaime Spencer, Richard Bowden, Simon Hadfield
2020 arXiv   pre-print
In the current monocular depth research, the dominant approach is to employ unsupervised training on large datasets, driven by warped photometric consistency.  ...  We show that within a single domain, our technique is comparable to both the current state of the art in monocular depth estimation and supervised feature representation learning.  ...  In the case of depth estimation, this is usually due to the assumption of photometric consistency, which starts to break down in dimly-lit environments.  ... 
arXiv:2003.13446v1 fatcat:i5eic7zz7zebdjozvygcd63nai

Approaches, Challenges, and Applications for Deep Visual Odometry: Toward to Complicated and Emerging Areas [article]

Ke Wang, Sai Ma, Junlan Chen, Fan Ren
2020 arXiv   pre-print
and matching, pose estimation.  ...  Then, using the offered criteria as the uniform measurements, we detailedly evaluate and discuss how deep learning improves the performance of VO from the aspects of depth estimation, feature extraction  ...  the complex environment in the real world.  ... 
arXiv:2009.02672v1 fatcat:zdnwt4lpmvbiromtpcxhxxpjxa

Unsupervised Deep Learning-Based RGB-D Visual Odometry

Qiang Liu, Haidong Zhang, Yiming Xu, Li Wang
2020 Applied Sciences  
Our two main contributions are: (i) during network training and pose estimation, the depth images are fed into the network to form a dual-stream structure with the RGB images, and a dual-stream deep neural  ...  (ii) the system adopts an unsupervised end-to-end training method, thus the labor-intensive data labeling task is not required.  ...  [7] started a research based on an unsupervised framework and proposed a network for monocular depth estimation, which can also be used to output camera poses.  ... 
doi:10.3390/app10165426 fatcat:2hleqjowgvcpzmzngyauchxwnm

Unsupervised Learning of Depth and Camera Pose with Feature Map Warping

Ente Guo, Zhifeng Chen, Yanlin Zhou, Dapeng Oliver Wu
2021 Sensors  
Estimating the depth of image and egomotion of agent are important for autonomous and robot in understanding the surrounding environment and avoiding collision.  ...  Most existing unsupervised methods estimate depth and camera egomotion by minimizing photometric error between adjacent frames.  ...  Therefore, in order to deal with the complex outdoor environment, real outdoor robotic applications focus on multiple sensor fusion.  ... 
doi:10.3390/s21030923 pmid:33573136 fatcat:3dl4zs7scbbvjcxb7blsy6jx4y

Multi-view Monocular Depth and Uncertainty Prediction with Deep SfM in Dynamic Environments [article]

Christian Homeyer, Oliver Lange, Christoph Schnörr
2022 arXiv   pre-print
3D reconstruction of depth and motion from monocular video in dynamic environments is a highly ill-posed problem due to scale ambiguities when projecting to the 2D image domain.  ...  In this work, we investigate the performance of the current State-of-the-Art (SotA) deep multi-view systems in such environments.  ...  Ranftl, R., Vineet, V., Chen, Q., Koltun, V.: Dense monocular depth estimation in complex dynamic scenes.  ... 
arXiv:2201.08633v1 fatcat:fnangw6thvafhpyuuycvy4ldiu

Unsupervised Monocular Depth Prediction for Indoor Continuous Video Streams [article]

Yinglong Feng, Shuncheng Wu, Okan Köpüklü, Xueyang Kang, Federico Tombari
2019 arXiv   pre-print
This paper studies unsupervised monocular depth prediction problem.  ...  Most of existing unsupervised depth prediction algorithms are developed for outdoor scenarios, while the depth prediction work in the indoor environment is still very scarce to our knowledge.  ...  The depth estimation in a monocular video stream highly relies on the motion between two adjacent frames of the camera.  ... 
arXiv:1911.08995v1 fatcat:aqc6zqapwbaepmnebqa7nygkwm

SelfVIO: Self-Supervised Deep Monocular Visual-Inertial Odometry and Depth Estimation [article]

Yasin Almalioglu, Mehmet Turan, Alp Eren Sari, Muhamad Risqi U. Saputra, Pedro P. B. de Gusmão, Andrew Markham, Niki Trigoni
2020 arXiv   pre-print
SelfVIO learns to jointly estimate 6 degrees-of-freedom (6-DoF) ego-motion and a depth map of the scene from unlabeled monocular RGB image sequences and inertial measurement unit (IMU) readings.  ...  Detailed comparisons prove that SelfVIO outperforms state-of-the-art VIO approaches in terms of pose estimation and depth recovery, making it a promising approach among existing methods in the literature  ...  Comparison of unsupervised monocular depth estimation between SfM-Learner [23] , CC [69] and the proposed SelfVIO.  ... 
arXiv:1911.09968v2 fatcat:vxucv3n6mred3p6pnh5w4wrvki
« Previous Showing results 1 — 15 out of 833 results