NumlockUknowSth commited on
Commit
cd6aef2
Β·
verified Β·
1 Parent(s): 7819a93

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +75 -3
README.md CHANGED
@@ -1,3 +1,75 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ ---
4
+
5
+
6
+ <div align="center">
7
+
8
+ <h1>ShotDirector: Directorially Controllable Multi-Shot Video Generation with Cinematographic Transitions</h1>
9
+
10
+ [![](https://img.shields.io/static/v1?label=ShotDirector&message=Project&color=purple)](https://github.com/UknowSth/ShotDirector)   [![](https://img.shields.io/static/v1?label=Paper&message=Arxiv&color=red&logo=arxiv)](https://arxiv.org/abs/2512.10286)   [![](https://img.shields.io/static/v1?label=Code&message=Github&color=blue&logo=github)](https://github.com/UknowSth/ShotDirector)    
11
+
12
+
13
+ <p><a href="https://scholar.google.com/citations?hl=zh-CN&user=TbZZSVgAAAAJ">Xiaoxue Wu</a><sup>1,2*</sup>,
14
+ <a href="https://scholar.google.com/citations?user=3fWSC8YAAAAJ">Xinyuan Chen</a><sup>2&dagger;</sup>,
15
+ <a href="https://wyhsirius.github.io/">Yaohui Wang</a><sup>2&dagger;</sup>,
16
+ <a href="https://scholar.google.com.hk/citations?user=gFtI-8QAAAAJ&amp;hl=zh-CN">Yu Qiao</a><sup>2&dagger;</sup>,
17
+ </p>
18
+
19
+
20
+ <span class="author-block"><sup>1</sup>Fudan University</span>
21
+ <span class="author-block"><sup>2</sup>Shanghai Artificial Intelligence Laboratory</span>
22
+
23
+
24
+ <span class="author-block"><sup>*</sup>Work done during internship at Shanghai AI Laboratory</span> <span class="author-block"><sup>&dagger;</sup>Corresponding author</span>
25
+
26
+ </div>
27
+
28
+ ## πŸ“₯ Installation
29
+ 1. Clone the Repository
30
+ ```
31
+ git clone https://github.com/UknowSth/ShotDirector.git
32
+ cd ShotDirector
33
+ ```
34
+ 2. Set up Environment
35
+ ```
36
+ conda create -n shotdirector python==3.11.9
37
+ conda activate shotdirector
38
+
39
+ pip install torch==2.5.1 torchvision==0.20.1 --index-url https://download.pytorch.org/whl/cu118
40
+ pip install -r requirements.txt
41
+ ```
42
+
43
+ ## πŸ€— Checkpoint
44
+
45
+ ### CineTrans-DiT
46
+ Download the weights of [Wan2.1-T2V-1.3B](https://huggingface.co/Wan-AI/Wan2.1-T2V-1.3B) and the weights required for Shotdirector. Place them in the `.ckpt/` folder as shown in the following diagram.
47
+
48
+ ```
49
+ ckpt/
50
+ │── Wan2.1/Wan2.1-T2V-1.3B/
51
+ β”‚ β”œβ”€β”€ config.json
52
+ β”‚ β”œβ”€β”€ diffusion_pytorch_model.safetensors
53
+ β”‚ β”œβ”€β”€ google/
54
+ β”‚ │── models_t5_umt5-xxl-enc-bf16.pth
55
+ β”‚ └── Wan2.1_VAE.pth
56
+ │── encoder.pt
57
+ │── model.pt
58
+ │── trans.pt
59
+ ```
60
+
61
+ For more inference details, please refer to our [GitHub repository](https://github.com/UknowSth/ShotDirector).
62
+
63
+ ## πŸ“‘ BiTeX
64
+ If you find [ShotDirector](https://github.com/UknowSth/ShotDirector.git) useful for your research and applications, please cite using this BibTeX:
65
+ ```
66
+ @misc{wu2025shotdirectordirectoriallycontrollablemultishot,
67
+ title={ShotDirector: Directorially Controllable Multi-Shot Video Generation with Cinematographic Transitions},
68
+ author={Xiaoxue Wu and Xinyuan Chen and Yaohui Wang and Yu Qiao},
69
+ year={2025},
70
+ eprint={2512.10286},
71
+ archivePrefix={arXiv},
72
+ primaryClass={cs.CV},
73
+ url={https://arxiv.org/abs/2512.10286},
74
+ }
75
+ ```