Hongyu LIU

Ph.D. student at HKUST.
My research interest includes 2D/3D Generation, digital human, neural rendering and contrastive learning.
I love fishing very much !!!

[Hongyu LIU @Kumapower1]

Excited to share our latest work AvatarPointillist! 🚀 We propose an AutoRegressive framework for 4D Gaussian Avatarization, enabling high-quality, animatable digital humans from sparse data.
#CVPR
project page: https://t.co/rnxj0Dj5u8
code: https://t.co/3gjI5EOk44

[Hongyu LIU @Kumapower1]

@dohae_metown Thank you so much for your support! We will open-source the code as soon as possible. Please stay tuned!

[Hongyu LIU @Kumapower1]

Excited to share our latest work AvatarPointillist! 🚀 We propose an AutoRegressive framework for 4D Gaussian Avatarization, enabling high-quality, animatable digital humans from sparse data.
#CVPR
project page: https://t.co/rnxj0Dj5u8
code: https://t.co/3gjI5EOk44

[宇多田ヒカルSTAFF @hikki_staff]

Hikaru Utada
"Mine or Yours" Music Video
5.7 (Wed) 21:00 (JST) premiere

https://t.co/fPp1J8fQJo

#HikaruUtada
#MineorYours

[宇多田ヒカルSTAFF @hikki_staff]

Hikaru Utada
Mine or Yours Music Video
Behind the Scenes

https://t.co/IoaTwpSgKQ

[宇多田ヒカルSTAFF @hikki_staff]

Thank you! @amazonmusicjp

"Mine or Yours" AVAILABLE NOW
https://t.co/GwH4KSnU4o

#HikaruUtada

[宇多田ヒカルSTAFF @hikki_staff]

Thank you @SpotifyJP

"Mine or Yours" AVAILABLE NOW
https://t.co/jAGvoxpnQO

#HikaruUtada

[Hongyu LIU @Kumapower1]

🚀 Check out our demo — amazing results across different domains! 🌍✨
https://t.co/k17QDqmeSu

[Hongyu LIU @Kumapower1]

We introduce AvatarArtist: Open-Domain 4D Avatarization, a novel single-image to 4D avatar method for diverse styles & domains. Check out our project page and code for more details! 🚀🔥 #CVPR2025 #Avatar

Project page: https://t.co/aFUS8XYUfP
code: https://t.co/Fx5z0PxAiF

[Hongyu LIU @Kumapower1]

@_akhaliq @camenduru @raywzy1

[Hongyu LIU @Kumapower1]

Our paper AvatarArtist is now live on Hugging Face! 🎉
Try the demo: upload your own photo & generate a 4D avatar in different styles 🎭✨
So many fun ways to play — check it out now!👇
https://t.co/k17QDqmeSu

[Hongyu LIU @Kumapower1]

paper:
https://t.co/GJYUU3EX7E

[Hongyu LIU @Kumapower1]

@taubnerfelix Thanks！😊 Your CAP4D paper is very promising!

[AI Bites | YouTube Channel @ai_bites]

Follow-Your-Emoji is a diffusion-based framework for portrait animation which animates a reference portrait with target landmark sequences. The main challenge of portrait animation is to preserve the identity of the reference portrait and transfer the target expression to this portrait while maintaining temporal consistency and fidelity.

Paper: Follow-Your-Emoji: Fine-Controllable and Expressive Freestyle Portrait Animation
Link: https://t.co/xvWZjX5mBi
Project: https://t.co/T6UbtvabN8

#AI #AI美女 #LLMs #deeplearning #machinelearning #GenerativeAI

[AK @_akhaliq]

An Object is Worth 64x64 Pixels: Generating 3D Object via Image Diffusion

discuss: https://t.co/rLg75oxUz4

We introduce a new approach for generating realistic 3D models with UV maps through a representation termed "Object Images." This approach encapsulates surface geometry, appearance, and patch structures within a 64x64 pixel image, effectively converting complex 3D shapes into a more manageable 2D format. By doing so, we address the challenges of both geometric and semantic irregularity inherent in polygonal meshes. This method allows us to use image generation models, such as Diffusion Transformers, directly for 3D shape generation. Evaluated on the ABO dataset, our generated shapes with patch structures achieve point cloud FID comparable to recent 3D generative models, while naturally supporting PBR material generation.

[宇多田ヒカル @utadahikaru]

「何色でもない花」MVのプレミア公開観てくれたみんなありがとう！

撮影手法は企業秘密とのことなんだけど、合成じゃなくて私も振り付けしてくれたアオイヤマダさんもスタッフも撮影チームもみんな氷の��で滑ったり震えながら作った映像です。今までで一番体張った撮影だったかも😅

本当にきれいな場所で、足元の雪が虹色に見えた瞬間があって寒さで幻覚でも見えたのかと思ったら、雪面の46°ハロっていう現象だったみたい😳

[Hongyu LIU @Kumapower1]

We changed "a DSLR portrait of a young man with a muscular jawline, stubble bread" to "a DSLR portrait of a young man with a muscular jawline, stubble bread, he has happy expression"

[Hongyu LIU @Kumapower1]

HeadArtist: Text-conditioned 3D Head Generation with Self Score Distillation.
paper page: https://t.co/hebHQ6VK5K
Project page: https://t.co/mIbyqk95i5
We propose Self Score Distillation to generate and edit high-fidelity 3D heads under text guidance.

[Hongyu LIU @Kumapower1]

prompt: a DSLR portrait of a pretty female with blue makeup.

[Hongyu LIU @Kumapower1]

prompt: a head of Arthas Menethil.

Lich King!!!

[Hongyu LIU @Kumapower1]

prompt: a head of RX-0 Unicorn Gundam. "Mobile Suit Gundam Unicorn" is one of my favorite Japanese anime

[Hongyu LIU @Kumapower1]

prompt: a head of Saber in Fate Stay Night, 3D, anime