Video to video example?

[antithing]

Hi, and thank you for making this code available!

Is there any example code of video restyling/video to video inference?

Is it possible to pass depth maps in as guidance?

[yoavhacohen]

Hi! Currently you can use FlowEdit and similar techniques for restyling/video to video inference. We're working on depth maps and other guidance options.

[antithing]

Thanks! I want to work directly in code, not in comfyui, how would I do i+ v2v inference in this scenario?

Could you point me in the right direction?

Also, as this model is so fast, is it technically possible to run in real-time? Eg, run at video one frame at a time and restyle at the video frame rate?

Thanks again!

[antithing]

I found this for v2v, can I pass a reference image as well using the below? Would it be technically possible to run in a streaming scenario, on a live video feed?

Thanks again!

`import torch
from diffusers import LTXConditionPipeline, LTXLatentUpsamplePipeline
from diffusers.pipelines.ltx.pipeline_ltx_condition import LTXVideoCondition
from diffusers.utils import export_to_video, load_video

pipe = LTXConditionPipeline.from_pretrained("Lightricks/LTX-Video-0.9.7-dev", torch_dtype=torch.bfloat16)
pipe_upsample = LTXLatentUpsamplePipeline.from_pretrained("Lightricks/ltxv-spatial-upscaler-0.9.7", vae=pipe.vae, torch_dtype=torch.bfloat16)
pipe.to("cuda")
pipe_upsample.to("cuda")
pipe.vae.enable_tiling()

def round_to_nearest_resolution_acceptable_by_vae(height, width):
    height = height - (height % pipe.vae_spatial_compression_ratio)
    width = width - (width % pipe.vae_spatial_compression_ratio)
    return height, width

video = load_video(
    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cosmos/cosmos-video2world-input-vid.mp4"
)[:21]  # Use only the first 21 frames as conditioning
condition1 = LTXVideoCondition(video=video, frame_index=0)

prompt = "The video depicts a winding mountain road covered in snow, with a single vehicle traveling along it. The road is flanked by steep, rocky cliffs and sparse vegetation. The landscape is characterized by rugged terrain and a river visible in the distance. The scene captures the solitude and beauty of a winter drive through a mountainous region."
negative_prompt = "worst quality, inconsistent motion, blurry, jittery, distorted"
expected_height, expected_width = 768, 1152
downscale_factor = 2 / 3
num_frames = 161

# Part 1. Generate video at smaller resolution
downscaled_height, downscaled_width = int(expected_height * downscale_factor), int(expected_width * downscale_factor)
downscaled_height, downscaled_width = round_to_nearest_resolution_acceptable_by_vae(downscaled_height, downscaled_width)
latents = pipe(
    conditions=[condition1],
    prompt=prompt,
    negative_prompt=negative_prompt,
    width=downscaled_width,
    height=downscaled_height,
    num_frames=num_frames,
    num_inference_steps=30,
    generator=torch.Generator().manual_seed(0),
    output_type="latent",
).frames

# Part 2. Upscale generated video using latent upsampler with fewer inference steps
# The available latent upsampler upscales the height/width by 2x
upscaled_height, upscaled_width = downscaled_height * 2, downscaled_width * 2
upscaled_latents = pipe_upsample(
    latents=latents,
    output_type="latent"
).frames

# Part 3. Denoise the upscaled video with few steps to improve texture (optional, but recommended)
video = pipe(
    conditions=[condition1],
    prompt=prompt,
    negative_prompt=negative_prompt,
    width=upscaled_width,
    height=upscaled_height,
    num_frames=num_frames,
    denoise_strength=0.4,  # Effectively, 4 inference steps out of 10
    num_inference_steps=10,
    latents=upscaled_latents,
    decode_timestep=0.05,
    image_cond_noise_scale=0.025,
    generator=torch.Generator().manual_seed(0),
    output_type="pil",
).frames[0]

# Part 4. Downscale the video to the expected resolution
video = [frame.resize((expected_width, expected_height)) for frame in video]

export_to_video(video, "output.mp4", fps=24)
`

[StarsTesla]

@antithing Hi, I am interesting about this v2v, is any example that you has been running.

[antithing]

I couldn't find an answer. Please post here if you do