Annotations of Abnormal Symptoms in Endoscopic Images

SELAB@HCMUS

Introduction

In this page, we provided additional annotations of Abnormal Symptoms in Endoscopic Images, which is a part of our work in the paper: Trung-Hieu Hoang, Hai-Dang Nguyen, Viet-Anh Nguyen, Thanh-An Nguyen, Vinh-Tiep Nguyen, and Minh-Triet Tran. 2019. Enhancing Endoscopic Image Classification with Symptom Localization and Data Augmentation. In Proceedings of the 27th ACM International Conference on Multimedia (MM’19), October 21–25, 2019, Nice, France. ACM, New York, NY, USA, 5 pages. https://doi.org/10.1145/3343031.3356073
For the KVASIR dataset, please refer to this link:
Bounding boxes of the following classes dyed-lifted-polyps, dyed-resection-margins, instruments, polyp are annotated from images taken from the following Dataset:
- KVASIR dataset (v2)
- Development dataset of the The Biomedia ACM MM Grand Challenge 2019
Samples of the bounding boxes are visualized in the below figure:

The annotation files are organized the same as the structure of the image folder of the KVASIR v2 dataset and BioMedia Development dataset the file name format is as follows:

image_name.txt

In each file, there are n lines corresponding to n symptoms position found in the corresponding image with format:

x1 y1 x2 y2 class_name

In order to download our annotations, please refer to the following link:

If you find this useful in your research, please consider to cite our work:

@inproceedings{Hoang:2019:EEI:3343031.3356073,

 author = {Hoang, Trung-Hieu and Nguyen, Hai-Dang and Nguyen, Viet-Anh and Nguyen, Thanh-An and Nguyen, Vinh-Tiep and Tran, Minh-Triet},

 title = {Enhancing Endoscopic Image Classification with Symptom Localization and Data Augmentation},

 booktitle = {Proceedings of the 27th ACM International Conference on Multimedia},

 series = {MM '19},

 year = {2019},

 isbn = {978-1-4503-6889-6},

 location = {Nice, France},

 pages = {2578--2582},

 numpages = {5},

 url = {http://doi.acm.org/10.1145/3343031.3356073},

 doi = {10.1145/3343031.3356073},

 acmid = {3356073},

 publisher = {ACM},

 address = {New York, NY, USA},

 keywords = {anatomical landmarks detection, endoscopic image, image classification, object detection, symptoms localization},

If there is any issues related to this work, please contact the corresponding author as follow:

TRUNG-HIEU HOANG

hthieu@selab.hcmus.edu.vn

Report abuse