Skip to content
Snippets Groups Projects
Commit 00ff70ad authored by Kshitij Nikhal's avatar Kshitij Nikhal
Browse files

update README

parent 82784552
No related branches found
No related tags found
No related merge requests found
# Weakly Supervised Face and Whole Body Recognition in Turbulent Environments # Weakly Supervised Face and Whole Body Recognition in Turbulent Environments
* main.py: Launch binary.
* pipeline.py : Generic training pipeline to load datasets, freeze parameters, and train and evaluate network.
* datasets/BRS.py and datasets/lrfid.py: Dataset loaders to load the BRS ReID and LRFID dataset along with their training and testing splits.
* datasets/utils/transforms_image.py: Contains code to warp/unwarp the image based on predicted tilts (generated offline).
* losses/memory_bank_loss.py: Loss function based on memory bank (works with supervised + unsupervised + semi-supervised)
* evaluation/* : all evaluations utils and helper functions.
* models/resnet50.py: Truncated resnet50 model.
* models/head.py: The self-attention projection (transformation) head attached to the end of the resnet model.
* models/tilt_map_predictor.py: Encoder-decoder to predict tilts.
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment