Gen-L-Video: Long Video Generation via Temporal Co-Denoising

This repository is the official implementation of Gen-L-Video.

You might be interested in Gen-L^2, which works better.

Introduction

TL;DR: A universal methodology that extends short video diffusion models for efficient multi-text conditioned long video generation and editing.

Current methodologies for video generation and editing, while innovative, are often confined to extremely short videos (typically less than 24 frames) and are limited to a single text condition. These constraints significantly limit their applications given that real-world videos usually consist of multiple segments, each bearing different semantic information. To address this challenge, we introduce a novel paradigm dubbed as Gen-L-Video capable of extending off-the-shelf short video diffusion models for generating and editing videos comprising hundreds of frames with diverse semantic segments without introducing additional training, all while preserving content consistency.

Essentially, this procedure establishes an abstract long video generator and editor without necessitating any additional training, enabling the generation and editing of videos of any length using established short video generation and editing methodologies.

Setup

Clone the Repo

git clone https://github.com/G-U-N/Gen-L-Video
cd Gen-L-Video
# The repo might be too large to clone because many long gifs are over 100 M. Fork the repo, delete the statics, and then clone it. 

Install Environment via Anaconda

conda env create -f requirements.yml
conda activate glv
conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.6 -c pytorch -c nvidia

Install Xformers

# (Optional) Makes the build much faster
pip install ninja
# Set TORCH_CUDA_ARCH_LIST if running and building on different GPU types
pip install -v -U git+https://github.com/facebookresearch/xformers.git@main#egg=xformers
# (this can take dozens of minutes)

Install SAM and Grounding DINO

pip install git+https://github.com/facebookresearch/segment-anything.git
pip install git+https://github.com/IDEA-Research/GroundingDINO.git

or

git clone https://github.com/facebookresearch/segment-anything.git
cd segment-anything
pip install -e .
cd ..
# If you have a CUDA environment, please make sure the environment variable CUDA_HOME is set. 
# If the cuda version of the system conflicts with the cudatoolkit version, See: https://github.com/G-U-N/Gen-L-Video/discussions/7
git clone https://github.com/IDEA-Research/GroundingDINO.git
cd GroundingDINO
pip install -e .

Note that if you are using GPU clusters that the management node has no access to GPU resources, you should submit the pip install -e . to the computing node as a computing task when building the GroundingDINO. Otherwise, it will not support detection computing through GPU.

Download Pretrained Weights

Make sure git-lfs is available. See: https://github.com/git-lfs/git-lfs/blob/main/INSTALLING.md

bash scripts/download_pretrained_models.sh

After downloading them, you should specify the absolute/relative path of them in the config files.

If you download all the above pretrained weights in the folder weights , set the configs files as follows:

1. In configs/tuning-free-inpaint/girl-glass.yaml

sam_checkpoint: "weights/sam_vit_h_4b8939.pth"
groundingdino_checkpoint: "weights/groundingdino_swinb_cogcoor.pth"
controlnet_path: "weights/edit-anything-v0-3"

2. In one-shot-tuning.py, set

adapter_paths={
    "pose":"weights/T2I-Adapter/models/t2iadapter_openpose_sd14v1.pth",
    "sketch":"weights/T2I-Adapter/models/t2iadapter_sketch_sd14v1.pth",
    "seg": "weights/T2I-Adapter/models/t2iadapter_seg_sd14v1.pth",
    "depth":"weights/T2I-Adapter/models/t2iadapter_depth_sd14v1.pth",
    "canny":"weights/T2I-Adapter/models/t2iadapter_canny_sd14v1.pth"
}

3. In configs/one-shot-tuning/hike.yaml, set

pretrained_model_path: "weights/anything-v4.0"

Then all the other weights are able to be automatically downloaded through the API of Hugging Face.

For users who are unable to download weights automatically

Here is an additional instruction for installing and running grounding dino.

# Notice: If you use 'pip install git+https://github.com/IDEA-Research/GroundingDINO.git'
# You should modify GroundingDINO_SwinB_cfg.py in python site-packages directory
# e.g. ~/miniconda3/envs/glv/lib/python3.8/site-packages/groundingdino/config/GroundingDINO_SwinB_cfg.py
cd GroundingDINO/groundingdino/config/ 
vim GroundingDINO_SwinB_cfg.py

set

text_encoder_type = "[Your Path]/bert-base-uncased"

Then

vim GroundingDINO/groundingdino/util/get_tokenlizer.py

Set

def get_pretrained_language_model(text_encoder_type):
    if text_encoder_type == "bert-base-uncased" or text_encoder_type.split("/")[-1]=="bert-base-uncased":
        return BertModel.from_pretrained(text_encoder_type)
    if text_encoder_type == "roberta-base":
        return RobertaModel.from_pretrained(text_encoder_type)
    raise ValueError("Unknown text_encoder_type {}".format(text_encoder_type))

Now you should be able to run your Grounding DINO with pre-downloaded bert weights.

Get your own control videos

git clone https://github.com/lllyasviel/ControlNet.git
cd ControlNet
git checkout f4748e3
mv ../process_data.py .
python process_data.py --v_path=../data --t_path=../t_data --c_path=../c_data --fps=10

Inference

1. One-Shot Tuning Method

accelerate launch one-shot-tuning.py --control=[your control]

[your control] can be set as pose , depth, seg, sketch, canny.

pose and depth are recommended.

2. Tuning-Free Method for videos with smooth semantic changes.

accelerate launch tuning-free-mix.py

3. Tuning-Free Edit Anything in Videos.

accelerate launch tuning-free-inpaint.py

4. Long video generation with pretrained.

accelerate launch follow-your-pose-long.py

5. Tuning-Free Long Video2Video generation

# canny 
accelerate launch tuning-free-control.py --config=./configs/tuning-free-control/girl-glass.yaml 
# hed
accelerate launch tuning-free-control.py --config=./configs/tuning-free-control/girl.yaml

Comparisons

Results

Most of the results can be generated with a single RTX 3090.

Multi-Text Conditioned Long Video Generation

https://github.com/G-U-N/Gen-L-Video/assets/60997859/9b370894-708a-4ed2-a2ac-abfa93829ea6

This video containing clips bearing various semantic information.

Long Video with Smooth Semantic Changes

All the following videos are directly generated with the pretrained Stable Diffusion weight without additional training.

Videos with Smooth Semantic Changes
"A man is boating, village." → "A man is walking by, city, sunset."	"A jeep car is running on the beach, sunny.” → "a jeep car is running on the beach, night."	"Lion, Grass, Rainy." → "Cat, Grass, Sun."	"A man is skiing in the sea." → "A man is surfing in the snow."

Edit Anything in Video

All the following videos are directly generated with the pretrained Stable Diffusion weight without additional training.

Edit Anything in Videos
Source Video	Mask of Sunglasses	"Sunglasses" → "Pink Sunglasses"	"Sunglasses" → "Cyberpunk Sunglasses with Neon Lights"
Source Video	Mask of Man	"Man" → "Bat Man"	"Man" → "Iron Man"

Controllable Video

Controllable Video
Pose Control	"Iron Man is fighting in the snow."	"A Van Gogh style painting of a man dancing."	"A man is running in the fire."
Depth Control	"Dog in the sun.""	"Tiger in the sun."	"Girl in the sun."

Tuning-Free Long Video-to-Video Generation

Tuning-Free Long Video-to-Video Generation
"Girls."	"Girls wearing sunglasses."

Long Video Generation with Pretrained Short Text-to-Video Diffusion Model

All the following videos are directly generated with the pre-trained VideoCrafter without additional training.

Long Video Generation with Pretrained Short Text-to-Video Diffusion Model
"Astronaut riding a horse." (Isolated)	"Astronaut riding a horse." (Gen-L-Video)	"Astronaut riding a horse, Loving Vincent Style." (Isolated)	"Astronaut riding a horse, Loving Vincent Style." (Gen-L-Video)
"A monkey is drinking water." (Isolated)	"A monkey