Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

๐Ÿš€ Introducing ๐ˆ๐ƒ๐Œ-๐•๐“๐Ž๐ : A novel diffusion model for image-based virtual try-on! ๐Ÿ‘— ๐Ÿ˜ Improves garment fidelity and generates authentic visuals. ๐Ÿ”ง Uses two modules to encode garment semantics: visual encoder for high-level & parallel UNet for low-level. Links below๐Ÿ‘‡

104,492 Aufrufe โ€ข vor 2 Jahren โ€ขvia X (Twitter)

8 Kommentare

Profilbild von Gradio
Gradiovor 2 Jahren

๐Ÿ’กIDM-VTON provides good support for detailed textual prompts for garments & person images ๐ŸŽจCustomization method using person-garment image pairs significantly improves fidelity. ๐Ÿš€ Ready to create your own virtual try-on apps? Build with Gradio today!

Profilbild von Gradio
Gradiovor 2 Jahren

๐Ÿ“ŠIDM-VTON outperforms other diffusion-based & GAN-based approaches in preserving garment details ๐ŸŒŸLooks effective in real-world scenarios! ๐Ÿ”IDM-VTON Official Gradio demo has been released on @huggingface Spaces! ๐Ÿ”—Demo:

Profilbild von LatinoRevolution (rev/acc)
LatinoRevolution (rev/acc)vor 2 Jahren

lol, this is great

Profilbild von Mahmoud Ghulman
Mahmoud Ghulmanvor 2 Jahren

The name is genius ๐Ÿ˜…

Profilbild von Yuchen Jin
Yuchen Jinvor 2 Jahren

Very interesting work, but is this expected?

Profilbild von dinos
dinosvor 2 Jahren

@yacineMTB killer @dingboard_ feature

Profilbild von dweedify
dweedifyvor 2 Jahren

Pretty cool!

Profilbild von Aswanth achoo'z
Aswanth achoo'zvor 2 Jahren

๐Ÿ˜…lol

ร„hnliche Videos

MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model with Gradio demo local demo: This paper studies the human image animation task, which aims to generate a video of a certain reference identity following a particular motion sequence. Existing animation works typically employ the frame-warping technique to animate the reference image towards the target motion. Despite achieving reasonable results, these approaches face challenges in maintaining temporal consistency throughout the animation due to the lack of temporal modeling and poor preservation of reference identity. In this work, we introduce MagicAnimate, a diffusion-based framework that aims at enhancing temporal consistency, preserving reference image faithfully, and improving animation fidelity. To achieve this, we first develop a video diffusion model to encode temporal information. Second, to maintain the appearance coherence across frames, we introduce a novel appearance encoder to retain the intricate details of the reference image. Leveraging these two innovations, we further employ a simple video fusion technique to encourage smooth transitions for long video animation. Empirical results demonstrate the superiority of our method over baseline approaches on two benchmarks. Notably, our approach outperforms the strongest baseline by over 38% in terms of video fidelity on the challenging TikTok dancing dataset. Code and model will be made available.

AK

810,024 Aufrufe โ€ข vor 2 Jahren