Presentation by Sourish Ghosh, Andrew Saba, and Anish Bhattacharya, part of the Air Lab Summer School 2020.
Sessions list, overviews, and links to repos: https://theairlab.org/summer2020
This session is an overview of popular methods used for object detection and tracking. Platform tools used for neural network inference is also discussed.
Colab 1 (Test detectors): https://colab.research.google.com/github/tensorflow/hub/blob/master/examples/colab/object_detection.ipynb
Colab 2 (Implement ResNet): https://colab.research.google.com/drive/1aJ5t-54OTUL25NhQpuDEiEA2oh5U8n65?usp=sharing
Tracker example: https://bitbucket.org/castacks/tracker_tutorial_ws/
Outline:
0:00 - Intro
1:46 - Timeline of methods
2:50 - Image Classification (using AlexNet)
6:40 - Region Proposals
7:18 - Two-stage methods (R-CNN, Fast R-CNN, and Faster R-CNN)
14:41 - One-stage methods (YOLO, RetinaNet, CornerNet)
39:06 - DETR
46:50 - Summary of Object Detection
49:35 - Inference Platform Tools
54:01 - OpenVino
58:35 - TensorRT
1:03:07 - Object Tracking
1:07:57 - Correlation Filters and MOSSE
1:17:33 - Median Flow
1:20:50 - Tracking-Learning-Detection
1:27:28 - Conclusion
Air Lab Website: https://theairlab.org
Twitter: https://twitter.com/airlabcmu
Facebook: https://www.facebook.com/airlabcmu
Medium: https://medium.com/airlabcmu
