ComfyUI - Object Detection & Segmentation - Florence2 & SAM2

preview_player
Показать описание
Together, Florence2 and SAM2 enhance ComfyUI's capabilities in image masking by offering precise control and flexibility over image detection and segmentation.

In this video, we show how you can easily and accurately mask objects in your video using Segment Anything 2 and Florence2.

Florence-2 is to interpret simple text prompts to perform tasks like captioning, object detection, and segmentation.

Segment Anything Model 2 (SAM 2) is a continuation of the Segment Anything project by Meta AI, designed to enhance the capabilities of automated image segmentation. Segmentation is a fundamental task in computer vision where the goal is to partition an image into segments, usually corresponding to different objects or regions. SAM 2 builds upon the original SAM by improving its ability to handle diverse and complex images.

***********************************

Comfyui tutorial, Учебное пособие по Comfyui, Comfyui ट्यूटोरियल, Tutoriel Comfyui, Tutorial Comfyui, Comfyui 튜토리얼
Comfyui stable diffusion, Install comfyui, comfyui video, controlnet comfyui, comfyui animateddiff, comfyui sdxl, comfyui upscale, comfyui video to video, comfyui manager, comfyui inpainting, comfyui ipadapter, comfyui faceswap

***********************************

------------------------------------
🌍 SOCIAL 

------------------------------------

#ComfyUI
#SAM2
#Florence2
Рекомендации по теме
Комментарии
Автор

Is it possible to use the model to segment Stable Difusion / Midjourney deformed pictures? (multiple fingers, blurry flaces, etc)

antoniojoaocastrocostajuni
Автор

thx for that easy video! what if...what if the preview image from 'florence run' is ok, the person is detected correctly, but in the output it masks the skirt only, not the whole person? tried different values in 'max_new_tokens' & 'num_beams', tried 'woman', 'woman figure', 'female', 'female figure', 'figure', 'dancer'-> without any effect... different video is ok, but in one particular video it just stick with the skirt only ... :(

yngeneer
Автор

Its pretty useful ! How can we do multiple object at the same time ?

elifmiami
Автор

Thanks for sharing!
- Can you please show how to do it with the new Sam2 Point Editor?
- Also, how to EXPORT ONLY the final selected segment into a video without the background?

Thanks ahead! 🙏

MrDanINSANE
Автор

What are the minimal requirements? im using 4060ti (16gb gpu ram) and i got a not enough memory error

erans
Автор

Sir, please take a look at how to make game sprites. thank you for the good video

너의그날은역사와함께
Автор

how can we use this cloth swapping ? plzzz tell

abhijeetmaurya