All Projects → youngbin-ro → Homography-Based-MOTDT

youngbin-ro / Homography-Based-MOTDT

Licence: MIT license
MOTDT with Homography Matrix for Multi-Object Tracking

Programming Languages

python
139335 projects - #7 most used programming language
Cuda
1817 projects
c
50402 projects - #5 most used programming language

Projects that are alternatives of or similar to Homography-Based-MOTDT

UniTrack
[NeurIPS'21] Unified tracking framework with a single appearance model. It supports Single Object Tracking (SOT), Video Object Segmentation (VOS), Multi-Object Tracking (MOT), Multi-Object Tracking and Segmentation (MOTS), Pose Tracking, Video Instance Segmentation (VIS), and class-agnostic MOT (e.g. TAO dataset).
Stars: ✭ 293 (+1295.24%)
Mutual labels:  object-tracking, multi-object-tracking
multi-camera-pig-tracking
Official Implementation of "Tracking Grow-Finish Pigs Across Large Pens Using Multiple Cameras"
Stars: ✭ 25 (+19.05%)
Mutual labels:  multi-object-tracking, homography
ObjectTracking
Object Detection and Tracking based on C++ and OpenCV
Stars: ✭ 23 (+9.52%)
Mutual labels:  object-tracking
petyr
Affine Tranformation and Homography library for Python
Stars: ✭ 16 (-23.81%)
Mutual labels:  homography
kernelized correlation filters gpu
Real-time visual object tracking using correlations filters and deep learning
Stars: ✭ 27 (+28.57%)
Mutual labels:  object-tracking
Squot
Squeak Object Tracker - Version control for arbitrary objects, currently with Git storage
Stars: ✭ 45 (+114.29%)
Mutual labels:  object-tracking
Yolov5-Deepsort
最新版本yolov5+deepsort目标检测和追踪,能够显示目标类别,支持5.0版本可训练自己数据集
Stars: ✭ 201 (+857.14%)
Mutual labels:  object-tracking
mosse-object-tracking
This is the implementation of MOSSE tracking algorithm (correlation filter based).
Stars: ✭ 89 (+323.81%)
Mutual labels:  object-tracking
PeekingDuck
A modular framework built to simplify Computer Vision inference workloads.
Stars: ✭ 143 (+580.95%)
Mutual labels:  object-tracking
CSA
Official implementation of CVPR2020 Paper "Cooling-Shrinking Attack"
Stars: ✭ 48 (+128.57%)
Mutual labels:  object-tracking
Gocv
Go package for computer vision using OpenCV 4 and beyond.
Stars: ✭ 4,511 (+21380.95%)
Mutual labels:  object-tracking
SiamFC-tf
A TensorFlow implementation of the SiamFC tracker, use with your own camera and video, or integrate to your own project 实时物体追踪,封装API,可整合到自己的项目中
Stars: ✭ 22 (+4.76%)
Mutual labels:  object-tracking
object-tracking
Multiple Object Tracking System in Keras + (Detection Network - YOLO)
Stars: ✭ 89 (+323.81%)
Mutual labels:  object-tracking
multi object tracker
An optical flow and Kalman Filter based tracker
Stars: ✭ 31 (+47.62%)
Mutual labels:  object-tracking
OpenMultiTracker
Tracker that uses computer vision methods to track players on the hockey field
Stars: ✭ 18 (-14.29%)
Mutual labels:  object-tracking
siam-mot
SiamMOT: Siamese Multi-Object Tracking
Stars: ✭ 446 (+2023.81%)
Mutual labels:  multi-object-tracking
zero-shot-object-tracking
Object tracking implemented with the Roboflow Inference API, DeepSort, and OpenAI CLIP.
Stars: ✭ 242 (+1052.38%)
Mutual labels:  object-tracking
robotics-level-4
This repo contains projects created using TensorFlow-Lite on Raspberry Pi and Teachable Machine. AI and ML capabilities have been integrated with Robot's software.
Stars: ✭ 34 (+61.9%)
Mutual labels:  object-tracking
ACVR2017
An Innovative Salient Object Detection Using Center-Dark Channel Prior
Stars: ✭ 20 (-4.76%)
Mutual labels:  object-tracking
homog2d
C++ 2D geometry library, handles points, lines, polylines, planar transformations (and other primitives), using homogeneous coordinates. Provided with complete manual and samples.
Stars: ✭ 70 (+233.33%)
Mutual labels:  homography

Homography-Based-MOTDT


Overview

overview

  • STEP1: Estimate bounding box of frame t+1 from the current frame t through Kalman Filter
  • STEP2: Detect object at time t+1 using R-FCN
  • STEP3: Filter objects estimated in STEP1 and objects detected in STEP2 through Non-Maximum Suppression
  • STEP4: Calculate homography matrix from frame t and t+1
  • STEP5: Create candidates by linearly transforming the existing object at time t through homography matrix obtained in STEP4
  • STEP6: Allocate bounding box candidates from STEP3 and STEP5 to each object based on IOU and ReIE features.

Tracking Examples

MOT17 Dataset

MOTDT (original)

  • The original model cannot maintain the track ID of object 1 (turned to 101), which is covered by object 105

Homography Based MOTDT (proposed)

  • Ours maintains the track ID of object 1 and 89 even though they are obscured by object 161 carrying a green bag.

VisDrone Dataset

MOTDT (original)

  • The original model cannot maintain the track ID of object 427 (turned to 509) due to a sudden change in camera angle

Homography Based MOTDT (proposed)

  • Ours maintains the track ID of object 515 even though there is a sudden change in camera angle at the end of the clip

Results

MOT17 Dataset

Original Proposed
idf1 0.503 0.522
Mostly Tracked 59 70
Mostly Lost 151 152
False Positive 919 3,057
Num_Misses 28,580 26,781
Num_Switches 200 198
Num_Fragment 706 574
MOTA 0.428 0.421
MOTP 0.152 0.164

VisDrone Dataset

Original Proposed
idf1 0.547 0.579
Mostly Tracked 75 97
Mostly Lost 94 97
False Positive 725 3,064
Num_Misses 22,704 19,818
Num_Switches 504 386
Num_Fragment 1,604 806
MOTA 0.524 0.538
MOTP 0.094 0.116

Implications

  • There has been a clear trade-off between the original and proposed method
  • False Positive increased a lot with additional bounding boxes generated by Homography, while Mostly Tracked measure which means the tracking success in the 80% of whole frames improved
  • Additionally, number of misses and number of fragments decreased considerably because of supplementary bounding boxes
  • Tracking time increased enormously, which is main downside of proposed method
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].