Illumination Dataset

Citation Author(s):
Maryam
Haghighat
University of New South Wales
Submitted by:
Maryam Haghighat
Last updated:
Tue, 05/17/2022 - 22:17
DOI:
10.21227/hd90-7m11
Research Article Link:
Links:
License:
718 Views
Categories:
0
0 ratings - Please login to submit your rating.

Abstract 

The dataset includes sample indoor and outdoor sequences with considerable variations in illumination.

Instructions: 

Datasets "Mall", "Oslo-LinPro" and "SJSU" are from AMOS dataset available on "http://amos.cse.wustl.edu/dataset".

Datasets "scene1", "scene2", "scene6" and "scene11" are from Phos dataset available on "http://robotics.pme.duth.gr/phos2.html".

Dataset Building is captured by the author at a frame rate of one per 15 minutes. 

The datasets are used to evaluate illumination estimation and compensation approaches in the following projects:

- M. Haghighat, R. Mathew, D. Taubman, "Illumination Estimation and Compensation of Low Frame Rate Video Sequences for Wavelet-Based Video Compression", IEEE Transactions on Image Processing. 

- M. Haghighat, R. Mathew, A. Naman, S. Young, and D. Taubman, "Rate-distortion optimized illumination estimation for wavelet-based video coding”, in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

- M. Haghighat, R. Mathew, A. Naman, and D. Taubman, “Lifting-based illumination adaptive transform (LIAT) using mesh-based illumination modelling”, in 2017 IEEE International Conference on Image Processing (ICIP).