change README

README.md

@@ -1,14 +1,34 @@
 # Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis Implementation
+
 [![Open in Streamlit](https://static.streamlit.io/badges/streamlit_badge_black_white.svg)](https://huggingface.co/spaces/flax-community/DietNerf-Demo) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1etYeMTntw5mh3FvJv4Ubb7XUoTtt5J9G?usp=sharing)
 
 <p align="center"><img width="450" alt="스크린샷 2021-07-04 오후 4 11 51" src="https://user-images.githubusercontent.com/77657524/126361638-4aad58e8-4efb-4fc5-bf78-f53d03799e1e.png"></p>
 
-Welcome to Putting NeRF on a Diet Project! 
-This project is the Pytorch, JAX/Flax based code implementation of this paper [Putting NeRF on a Diet : Ajay Jain, Matthew Tancik, Pieter Abbeel, Arxiv : https://arxiv.org/abs/2104.00677] 
-The model generates the novel view synthesis redering (NeRF: Neural Radiances Field) with Fewshot learning scheme.
-The semantic loss use the pre-trained CLIP Vision Transformer embedding. This information can give a 2D supervision for 3D.
-The Diet NeRF result outperforms the original NeRF in 3D reconstruction and neural rendering with only few images. 
+This project attempted to implement the paper **[Putting NeRF on a Diet](https://arxiv.org/abs/2104.00677)** (DietNeRF) in JAX/Flax.
+DietNeRF is designed for rendering quality novel views in few-shot learning scheme, a task that vanilla NeRF (Neural Radiance Field) struggles.
+To achieve this, the author coins **Semantic Consistency Loss** to supervise DietNeRF by prior knowledge from CLIP Vision Transformer. Such supervision enables DietNeRF to learn 3D scene reconstruction with CLIP's prior knowledge on 2D views.  
+
+Besides this repo, you can check our write-up and demo here:
+- ✍️ **[Write-up in Notion](https://steep-cycle-f6b.notion.site/DietNeRF-Putting-NeRF-on-a-Diet-4aeddae95d054f1d91686f02bdb74745)**: more details of DietNeRF and our experiments
+- ✨ **[Demo in Hugging Face Space](https://huggingface.co/spaces/flax-community/DietNerf-Demo)**: showcase our trained DietNeRFs by Streamlit
+
+## 🤩 Demo
+1. You can check out [our demo in Hugging Face Space](https://huggingface.co/spaces/flax-community/DietNerf-Demo)
+2. Or you can set up our Streamlit demo locally (model checkpoints will be fetched automatically upon startup)
+```shell
+pip install -r requirements_demo.txt
+streamlit run app.py
+```
+
+<p align="center"><img width="600" height="400" alt="Streamlit Demo" src="assets/space_demo.png"></p>
+
+## ✨ Implementation
+
+Our code is written in JAX/ Flax and mainly based upon [jaxnerf](https://github.com/google-research/google-research/tree/master/jaxnerf) from Google Research. The base code is highly optimized in GPU & TPU. For semantic consistency loss, we utilize pretrained CLIP Vision Transformer from [transformers](https://github.com/huggingface/transformers) library.  
 
+To learn more about DietNeRF, our experiments and implementation, you are highly recommended to check out our very detailed **[Notion write-up](https://www.notion.so/DietNeRF-Putting-NeRF-on-a-Diet-4aeddae95d054f1d91686f02bdb74745)**!
+
+<p align="center"><img width="500" height="600" alt="스크린샷 2021-07-04 오후 4 11 51" src="assets/report_thumbnail.png"></p> 
 
 ##  🤗 Hugging Face Hub Repo URL:
 We will also upload our project on the Hugging Face Hub Repository Also. 
@@ -49,23 +69,6 @@ Our JAX/Flax implementation currently supports:
 </tbody>
 </table>
 
-## 🤩 Demo
-
-- Streamlit Space Demo
-
-You can check our Streamlit Space demo on following site !
-With any input camera pose, we can render the novel view synthesis.
-[https://huggingface.co/spaces/flax-community/DietNerf-Demo](https://huggingface.co/spaces/flax-community/DietNerf-Demo)
-
-- Colab Demo
-
-Moreover, we prapare the colab ipython notebook for you.
-You need colab pro account for running our model on the colab(For memory issue)
-[https://colab.research.google.com/drive/1etYeMTntw5mh3FvJv4Ubb7XUoTtt5J9G?usp=sharing
-](https://colab.research.google.com/drive/1etYeMTntw5mh3FvJv4Ubb7XUoTtt5J9G?usp=sharing
-)
-
-
 ## 💻 Installation
 
 ```bash