make folders for IEEE Access

kktsubota · Feb 27, 2023 · 7fe35ed · 7fe35ed
1 parent 34c8375
commit 7fe35ed
Show file tree

Hide file tree

Showing 9 changed files with 55 additions and 49 deletions.
diff --git a/Access23/README.md b/Access23/README.md
@@ -0,0 +1,2 @@
+# Comprehensive Comparisons of Uniform Quantization in Deep Image Compression
+Official implementation of "Comprehensive Comparisons of Uniform Quantization in Deep Image Compression" in IEEE Access 2023.
diff --git a/LICENSE → ICIP21/LICENSE b/LICENSE → ICIP21/LICENSE
diff --git a/Pipfile → ICIP21/Pipfile b/Pipfile → ICIP21/Pipfile
diff --git a/Pipfile.lock → ICIP21/Pipfile.lock b/Pipfile.lock → ICIP21/Pipfile.lock
diff --git a/ICIP21/README.md b/ICIP21/README.md
@@ -0,0 +1,51 @@
+# Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression
+Official implementation of "Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression" in ICIP 2021.
+
+## Dataset
+Prepare ImageNet and the Kodak dataset.
+
+We removed images whose shorter edge is smaller than 256 pixels for ImageNet.
+
+Download the list of images that we used in our experiments by the following commands.
+
+```
+mkdir datasets
+wget https://github.com/kktsubota/uniform-quantizers/releases/download/pre/ImageNet256.txt -O datasets/ImageNet256.txt
+```
+
+## Environment
+* CUDA==10.0
+* CUDNN==7.6.0
+* Python
+
+```
+pip install pipenv
+pipenv install
+```
+
+Refer to `Pipfile` if you download packages manually.
+
+## Usage
+
+```bash
+# train a model
+python main.py --verbose --checkpoint_dir checkpoints/l0.01_aun_aun --qua_ent AUN-Q train --lambda 0.01 --qua_dec AUN-Q --train_root /path/to/ImageNet/train/
+
+# evaluate the model
+python evaluate.py /path/to/Kodak/images/ --qua_ent AUN-Q --checkpoint_dir checkpoints/l0.01_aun_aun/
+```
+
+You can train other combinations of approximation methods by specifying `--qua_ent` and `--qua_dec`.
+
+Please select from `{AUN-Q, STE-Q, U-Q, SGA-Q}` for each option.
+
+## Citation
+```
+@inproceedings{tsubotaICIP21,
+    title = {Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression},
+    author = {Tsubota, Koki and Aizawa, Kiyoharu},
+    booktitle = {ICIP},
+    year = {2021},
+    pages={2089-2093}
+}
+```
diff --git a/evaluate.py → ICIP21/evaluate.py b/evaluate.py → ICIP21/evaluate.py
diff --git a/main.py → ICIP21/main.py b/main.py → ICIP21/main.py
diff --git a/module.py → ICIP21/module.py b/module.py → ICIP21/module.py
diff --git a/README.md b/README.md
@@ -1,50 +1,3 @@
-# uniform-quantizers
-Official implementation of "Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression".
+# Uniform-Quantizers
+This repository provides the source code of our two papers: "Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression" in ICIP 2021 and "Comprehensive Comparisons of Uniform Quantization in Deep Image Compression" in IEEE Access 2023.
 
-## Dataset
-Prepare ImageNet and the Kodak dataset.
-
-We removed images whose shorter edge is smaller than 256 pixels for ImageNet.
-
-Download the list of images that we used in our experiments by the following commands.
-
-```
-mkdir datasets
-wget https://github.com/kktsubota/uniform-quantizers/releases/download/pre/ImageNet256.txt -O datasets/ImageNet256.txt
-```
-
-## Environment
-* CUDA==10.0
-* CUDNN==7.6.0
-* Python
-
-```
-pip install pipenv
-pipenv install
-```
-
-Refer to `Pipfile` if you download packages manually.
-
-## Usage
-
-```bash
-# train a model
-python main.py --verbose --checkpoint_dir checkpoints/l0.01_aun_aun --qua_ent AUN-Q train --lambda 0.01 --qua_dec AUN-Q --train_root /path/to/ImageNet/train/
-
-# evaluate the model
-python evaluate.py /path/to/Kodak/images/ --qua_ent AUN-Q --checkpoint_dir checkpoints/l0.01_aun_aun/
-```
-
-You can train other combinations of approximation methods by specifying `--qua_ent` and `--qua_dec`.
-
-Please select from `{AUN-Q, STE-Q, U-Q, SGA-Q}` for each option.
-
-## Citation
-```
-@inproceedings{tsubotaICIP21,
-    title = {Comprehensive Comparisons of Uniform Quantizers for Deep Image Compression},
-    author = {Tsubota, Koki and Aizawa, Kiyoharu},
-    booktitle = {ICIP},
-    year = {2021}
-}
-```
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		# Comprehensive Comparisons of Uniform Quantization in Deep Image Compression
		Official implementation of "Comprehensive Comparisons of Uniform Quantization in Deep Image Compression" in IEEE Access 2023.