Update README.md

update readme
seasonSH · Jun 11, 2022 · 6ffd7c9 · 6ffd7c9
1 parent 4709420
commit 6ffd7c9
Showing 1 changed file with 78 additions and 78 deletions.
diff --git a/README.md b/README.md
@@ -23,14 +23,14 @@ Official Implementation of our SemanticStyleGAN paper for training and inference
 # Table of Contents
 - [Installation](#installation)
 - [Pretrained Models](#pretrained-models)
+- [Inference](#inference)
+  * [Synthesis](#synthesis)
+  * [Inversion](#inversion)
+  * [Computing Metrics](#computing-metrics)
 - [Training](#training)
   * [Data Preparation](#data-preparation)
   * [Training SemanticStyleGAN](#training-semanticstylegan)
-- [Inference](#inference)
-  * [Computing Metrics](#computing-metrics)
-  * [Inversion](#inversion)
-  * [Synthesis](#synthesis)
-- [Domain Adaptation](#domain-adaptation)
+  * [Domain Adaptation](#domain-adaptation)
 - [Credits](#credits)
 - [Acknowledgments](#acknowledgments)
 - [Citation](#citation)
@@ -56,6 +56,78 @@ In this repository, we provide pretrained models for various domains.
 
 <br>
 
+# Inference
+
+## Synthesis
+### Random Synthesis
+In `visualize/generate.py`, we provide a script for sampling random images and their corresponding segmentation masks with SemanticStyleGAN.
+An example command is provided below:
+```
+python visualize/generate.py \
+/path/to/checkpoint \
+--outdir results/samples \
+--sample 20 \
+--save_latent
+```
+The `save_latent` flag will save the w latent code of each synthesized image in a separate `.npy` file.
+
+### Local Latent Interpolation
+<img src="docs/local_interpolation.gif" width="400px"/>  
+
+In `visualize/generate_video.py`, we provide a script for visualizing the local interpolation by SemanticStyleGAN.
+An example command is provided below:  
+```
+python visualize/generate_video.py \
+/path/to/checkpoint \
+--outdir /results/interpolation \
+--latent /results/samples/000000_latent.npy
+```
+Here, `/results/samples/000000_latent.npy` is the latent code either generated by `visualize/generate.py` or output by `visualize/invert.py`. You can also ignore the `--latent` flag for 
+generating a video with a random latent code. The scripts will create sevaral mp4 files udner the output folder, each shows the interpolation animation in
+a specific latent subspace.
+
+### Synthesizing Components
+In `visualize/generate_components.py`, we provide a script for visualizing the components synthesized by SemanticStyleGAN, where we will gradually add more local generators
+to the synthesis procedure.
+An example command is provided below:
+```
+python visualize/generate_components.py \
+/path/to/checkpoint \
+--outdir /results/components \
+```
+This command will synthesize random images in a gradual manner. You can also visualize the components of a specific target image with the following command:
+```
+python visualize/generate_components.py \
+/path/to/checkpoint \
+--outdir /results/components/000000 \
+--latent /results/samples/000000_latent.npy \
+```
+
+## Computing Metrics
+Given a trained generator and a prepared inception file, we can compute the metrics with following command:
+```
+python calc_fid.py \
+--ckpt /path/to/checkpoint \
+--inception /path/to/inception/file 
+```
+
+## Inversion
+### Optimization-based
+You can use `visualize/invert.py` for inverting real images into the latent space of SemanticStyleGAN via optimization:
+```
+python visualize/invert.py \
+--ckpt /path/to/checkpoint \
+--imgdir /path/to/img/directory/ \
+--outdir /path/to/output/directory/ \
+--size 512 \
+```
+This script will save the reconstructed images and their corresponding w-plus latent codes in separate sub-directories under the `outdir`. Additionally, you can set `--finetune_step` to a non-zero integer for pivotal tuning inversion.
+
+### Encoder
+TODO
+
+<br>
+
 # Training
 
 ## Data Preparation
@@ -116,79 +188,7 @@ Additionally, if you have tensorboard installed, you can visualize tensorboard l
 
 <br>
 
-# Inference
-
-## Computing Metrics
-Given a trained generator and a prepared inception file, we can compute the metrics with following command:
-```
-python calc_fid.py \
---ckpt /path/to/checkpoint \
---inception /path/to/inception/file 
-```
-
-## Inversion
-### Optimization-based
-You can use `visualize/invert.py` for inverting real images into the latent space of SemanticStyleGAN via optimization:
-```
-python visualize/invert.py \
---ckpt /path/to/checkpoint \
---imgdir /path/to/img/directory/ \
---outdir /path/to/output/directory/ \
---size 512 \
-```
-This script will save the reconstructed images and their corresponding w-plus latent codes in separate sub-directories under the `outdir`. Additionally, you can set `--finetune_step` to a non-zero integer for pivotal tuning inversion.
-
-### Encoder
-TODO
-
-## Synthesis
-### Random Synthesis
-In `visualize/generate.py`, we provide a script for sampling random images and their corresponding segmentation masks with SemanticStyleGAN.
-An example command is provided below:
-```
-python visualize/generate.py \
-/path/to/checkpoint \
---outdir results/samples \
---sample 20 \
---save_latent
-```
-The `save_latent` flag will save the w latent code of each synthesized image in a separate `.npy` file.
-
-### Local Latent Interpolation
-<img src="docs/local_interpolation.gif" width="400px"/>  
-
-In `visualize/generate_video.py`, we provide a script for visualizing the local interpolation by SemanticStyleGAN.
-An example command is provided below:  
-```
-python visualize/generate_video.py \
-/path/to/checkpoint \
---outdir /results/interpolation \
---latent /results/samples/000000_latent.npy
-```
-Here, `/results/samples/000000_latent.npy` is the latent code either generated by `visualize/generate.py` or output by `visualize/invert.py`. You can also ignore the `--latent` flag for 
-generating a video with a random latent code. The scripts will create sevaral mp4 files udner the output folder, each shows the interpolation animation in
-a specific latent subspace.
-
-### Synthesizing Components
-In `visualize/generate_components.py`, we provide a script for visualizing the components synthesized by SemanticStyleGAN, where we will gradually add more local generators
-to the synthesis procedure.
-An example command is provided below:
-```
-python visualize/generate_components.py \
-/path/to/checkpoint \
---outdir /results/components \
-```
-This command will synthesize random images in a gradual manner. You can also visualize the components of a specific target image with the following command:
-```
-python visualize/generate_components.py \
-/path/to/checkpoint \
---outdir /results/components/000000 \
---latent /results/samples/000000_latent.npy \
-```
-
-<br>
-
-# Domain Adaptation
+## Domain Adaptation
 
 In `train_adaptation.py`, we provide a script for performing domain adaptation on image-only datasets.
 To do this, you first need to create an LMDB for the target image dataset.