Posterior Sampling Using Diffusion Models for HDR Reconstruction

Project code for CS 231N Spring 2025

Author: Jamin Xie

Paper and Poster:

Paper: PS_HDR.pdf

Poster: PS_HDR_Poster.pdf

All files configs/task/hdr*.yaml are configuration files for new forward operators
forward_operator/__init__.py modified to include the new forward operators
baseline_reconstruction.py contains logic for running baseline
baseline.sh runs the baseline in command line
compress_figures.py compresses figures to pdf
make_figures.py makes figures
plot_ablation.py makes the ablation study graph
Updated version of requirements.txt with fixed dependencies is in environment.yml
Full run history to make figures in paper and poster are in samples.sh

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
README.assets		README.assets
commands		commands
configs		configs
cores		cores
dataset		dataset
figures		figures
forward_operator		forward_operator
model		model
scripts		scripts
tables		tables
.gitignore		.gitignore
LICENSE		LICENSE
PS_HDR.pdf		PS_HDR.pdf
PS_HDR_Poster.pdf		PS_HDR_Poster.pdf
README.md		README.md
README_DAPS.md		README_DAPS.md
baseline.sh		baseline.sh
baseline_reconstruction.py		baseline_reconstruction.py
compress_figures.py		compress_figures.py
data.py		data.py
download.sh		download.sh
environment.yml		environment.yml
eval.py		eval.py
evaluate_fid.py		evaluate_fid.py
make_figures.py		make_figures.py
plot_ablation.py		plot_ablation.py
posterior_sample.py		posterior_sample.py
requirements.txt		requirements.txt
sampler.py		sampler.py
samples.sh		samples.sh
simple_samples.sh		simple_samples.sh