Semi-Automatic Annotation For Visual Object Tracking

We propose a semi-automatic bounding box annotation method for visual object tracking by utilizing temporal information with a tracking-by-detection approach. For detection, we use an off-the-shelf object detector which is trained iteratively with the annotations generated by the proposed method, an...

Full description

Saved in:
Bibliographic Details
Main Authors: Ince, Kutalmis Gokalp, Koksal, Aybora, Fazla, Arda, Alatan, A. Aydin
Format: Conference Proceeding
Language:English
Subjects:
Annotations
Conferences
Detectors
Manuals
Object detection
Training
Visualization
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a semi-automatic bounding box annotation method for visual object tracking by utilizing temporal information with a tracking-by-detection approach. For detection, we use an off-the-shelf object detector which is trained iteratively with the annotations generated by the proposed method, and we perform object detection on each frame independently. We employ Multiple Hypothesis Tracking (MHT) to exploit temporal information and to reduce the number of false-positives which makes it possible to use lower objectness thresholds for detection to increase recall. The tracklets formed by MHT are evaluated by human operators to enlarge the training set. This novel incremental learning approach helps to perform annotation iteratively. The experiments performed on AUTH Multidrone Dataset reveal that the annotation workload can be reduced up to 96% by the proposed approach. Resulting uav_detection_2 annotations and our codes are publicly available at github.com/aybora/Semi-Automatic-Video-Annotation-OGAM.
ISSN:2473-9944
DOI:10.1109/ICCVW54120.2021.00143