scholarly journals NM-Net: Mining Reliable Neighbors for Robust Feature Correspondences

2019 ◽  
Author(s):  
Chen Zhao ◽  
Zhiguo Cao ◽  
Chi Li ◽  
Xin Li ◽  
Jiaqi Yang
Keyword(s):  
Feature Correspondences

Multicamera Video Stitching for Multiple Human Tracking

2021 ◽  
Vol 11 (1) ◽  
pp. 17-38
Author(s):  
S. Vasuhi ◽  
A. Samydurai ◽  
Vijayakumar M.
Keyword(s):  
Fuzzy Inference ◽  
Human Tracking ◽  
Foreground Extraction ◽  
Inference System ◽  
Video Stitching ◽  
Novel Approach ◽  
Key Points ◽  
Feature Correspondences ◽  
Centroid Point ◽  
Stitching Techniques

In this paper, a novel approach is proposed to track humans for video surveillance using multiple cameras and video stitching techniques. SIFT key points are extracted from all camera inputs. Using k-d tree algorithm, all the key points are matched and random sample consensus (RANSAC) is used to identify the match correspondence among all the matched points. Homography matrix is calculated using four matched robust feature correspondences, the images are warped with respect to the other images, and the human tracking is performed on the stitched image. To identify the human in the stitched video, background modeling is performed using fuzzy inference system and perform foreground extraction. After foreground extraction, the blobs are constructed around each detected human and centroid point is calculated for each blob. Finally, tracking of multiple humans is done by Kalman filter (KF) with Hungarian algorithm.

scholarly journals Lost in the Woods? Place Recognition for Navigation in Difficult Forest Environments

2020 ◽  
Vol 7 ◽  
Author(s):  
James Garforth ◽  
Barbara Webb
Keyword(s):  
State Of The Art ◽  
Robotic Systems ◽  
Place Recognition ◽  
Challenging Problem ◽  
Data Set ◽  
Loop Closure ◽  
Urban Scenes ◽  
Feature Correspondences ◽  
Recognition Systems ◽  
Crucial Part

Forests present one of the most challenging environments for computer vision due to traits, such as complex texture, rapidly changing lighting, and high dynamicity. Loop closure by place recognition is a crucial part of successfully deploying robotic systems to map forests for the purpose of automating conservation. Modern CNN-based place recognition systems like NetVLAD have reported promising results, but the datasets used to train and test them are primarily of urban scenes. In this paper, we investigate how well NetVLAD generalizes to forest environments and find that it out performs state of the art loop closure approaches. Finally, integrating NetVLAD with ORBSLAM2 and evaluating on a novel forest data set, we find that, although suitable locations for loop closure can be identified, the SLAM system is unable to resolve matched places with feature correspondences. We discuss additional considerations to be addressed in future to deal with this challenging problem.

scholarly journals Removal of specular reflections from image sequences using feature correspondences

2017 ◽  
Vol 28 (3-4) ◽  
pp. 409-420 ◽  
Author(s):  
Syed. M. Z. Abbas Shah ◽  
Stephen Marshall ◽  
Paul Murray
Keyword(s):  
Image Sequences ◽  
Specular Reflections ◽  
Feature Correspondences

Multi-waypoint visual homing in piecewise linear trajectory

Robotica ◽  
2012 ◽  
Vol 31 (3) ◽  
pp. 479-491 ◽  
Author(s):  
Yu Fu ◽  
Tien-Ruey Hsiang ◽  
Sheng-Luen Chung
Keyword(s):  
Piecewise Linear ◽  
Image Sequence ◽  
Linear Trajectory ◽  
Visual Homing ◽  
Feature Correspondences ◽  
Vertical Displacements ◽  
Office Environments ◽  
Navigation Method

SUMMARYThis paper proposes an image sequence-based navigation method under the teaching-replay framework for robots in piecewise linear routes. Waypoints used by the robot contain either the positions with large heading changes or selected midway positions between junctions. The robot applies local visual homing to move between consecutive waypoints. The arrival at a waypoint is determined by minimizing the average vertical displacements of feature correspondences. The performance of the proposed approach is supported by extensive experiments in hallway and office environments. While the homing speed of robots using other approaches is constrained by the speed in the teaching phase, our robot is not bounded by such limit and can travel much faster without compromising the homing accuracy.

Determination of feature correspondences in stereo images using a calibration polygon

1997 ◽  
Vol 30 (9) ◽  
pp. 1387-1400 ◽  
Author(s):  
Jui-Man Chiu ◽  
Zen Chen ◽  
Jen-Hui Chuang ◽  
Tsorng-Lin Chia
Keyword(s):  
Stereo Images ◽  
Feature Correspondences
Sign in / Sign up

Export Citation Format

Share Document