mehmetkeremturkcan/Suturing-LTX-T2V

Towards Suturing World Models (LTX-Video, t2v)

This repository hosts the fine-tuned LTX-Video text-to-video (t2v) diffusion model specialized for generating realistic robotic surgical suturing videos, capturing fine-grained sub-stitch actions including needle positioning, targeting, driving, and withdrawal. The model can differentiate between ideal and non-ideal surgical techniques, making it suitable for applications in surgical training, skill evaluation, and autonomous surgical system development.

Model Details

• Base Model: LTX-Video
• Resolution: 768×512 pixels (Adjustable)
• Frame Length: 49 frames per generated video (Adjustable)
• Fine-tuning Method: Low-Rank Adaptation (LoRA)
• Data Source: Annotated laparoscopic surgery exercise videos (∼2,000 clips)

Usage Example

import torch
from diffusers import LTXPipeline
from diffusers.utils import export_to_video

pipe = LTXPipeline.from_pretrained(
    "Lightricks/LTX-Video", torch_dtype=torch.bfloat16
).to("cuda")
pipe.load_lora_weights("mehmetkeremturkcan/Suturing-LTX-T2V", weight_name="pytorch_lora_weights.safetensors", adapter_name="ltxv-lora")
pipe.set_adapters(["ltxv-lora"], [1.])

for i in range(10):
    video = pipe("suturingv2 A needledrivingnonideal clip, generated from a backhand task.", height=512,
        width=768,
        num_frames=49,
        num_inference_steps=30,).frames[0]
    export_to_video(video, "ltx_lora_t2v_{}.mp4".format(str(i)), fps=8)

Applications

• Surgical Training: Generate demonstrations of both ideal and non-ideal surgical techniques for training purposes.
• Skill Evaluation: Assess surgical skills by comparing actual procedures against model-generated standards.
• Robotic Automation: Inform autonomous surgical robotic systems for real-time guidance and procedure automation.

Quantitative Performance

Metric	Performance
L2 Reconstruction Loss	0.32576
Inference Time	~6.1 seconds per video

Future Directions

Further improvements will focus on increasing model robustness, expanding the dataset diversity, and enhancing real-time applicability to robotic surgical scenarios.