Yihang Luo

Ph.D Student, Computer Vision @MMLabNTU | Ex Undergrad, SuperComputing Team @NTUsg

[Andrea Tagliasacchi @CVPR @taiyasaki]

This is why I refuse to work on camera estimation problems. For embodied AI, I consider it a **solved** problem at large – there are better things to spend your time on.

It's only a "problem" in academia.

[Xingang Pan @XingangP]

Introducing 𝗦𝗧𝗿𝗲𝗮𝗺𝟯𝗥, a new 3D geometric foundation model for efficient 3D reconstruction from streaming input. Similar to LLMs, STream3R uses casual attention during training and KVCache at inference.

No need to worry about post-alignment or reconstructing from scratch. You can easily add new frames and update the reconstruction incrementally. Great work by Yushi @GROS17121524 and Yihang @TheYihangLuo !

Project: https://t.co/ldAWpOhrZe
arXiv: https://t.co/I7H7QFzOiv
Code: https://t.co/9MHeGDV9CK

See a streaming reconstruction of our S-Lab lobby below!

[Yihang Luo @onelineluo]

STream3R reformulates dense 3D reconstruction into a sequential registration task with causal attention.

Just tried 3D reconstruction on a #GrokImagine video using #STream3R🫡!

Check out STream3R on our GitHub for more👨‍💻:
https://t.co/sj4fRe3vts

[Elon Musk @elonmusk]

Grok Imagine prompt:

A lone swordsman in a tattered cloak battles a massive sand serpent in a desert coliseum at midday, his blade flashing as dust clouds swirl around him. The arena is surrounded by crumbling stone pillars and a blazing sun overhead. Harsh sunlight cast...

[AK @_akhaliq]

STream3R

Scalable Sequential 3D Reconstruction with Causal Transformer

[Yihang Luo @onelineluo]

🔍 Explore our Code and HuggingFace Demo here:

💻Code: https://t.co/H3RxMWpY8Z
🌄Demo: https://t.co/yDVPP2Eo06

[Yihang Luo @onelineluo]

💥 Consistent Multi-View Diffusion for 3D Enhancement 💥

Introducing our work #3DEnhancer @CVPR: a multi-view diffusion model that enhances multi-view images to improve 3D models.

📰arXiv: https://t.co/eNvgSTsKWN
🔥Project: https://t.co/VDPG5NvRSt

[Yihang Luo @onelineluo]

@ShangchenZhou Wow, congrats!

[Chen Change Loy @ccloy]

🔬 Our study introduces "Upscale-A-Video," a text-guided latent diffusion framework for video upscaling. It ensures temporal coherence locally & globally, balancing fidelity and quality.

🚀 Project page: https://t.co/fYaPM8G69B
💻 GitHub: https://t.co/aWTs5lN60z
🎥 Video: https://t.co/Q1a2yzxl1u

@ShangchenZhou @peiqing001 @Iceclearwjy @TheYihangLuo

[Shangchen Zhou @ShangchenZhou]

Our #ProPainter👨‍🎨 @huggingface demo is finally live! Remove any object with just a click. Dive in and give it a try! 🛹

Access the demo here: https://t.co/AadzmGrxQw

[Yuekun Dai @YuekunDai]

🌟Do you feel these bright spots quite annoying in smartphone photography?
🌟Our highlight paper proposes the first learning-based method specialized for this problem. Come and check in this afternoon #CVPR2023 ! ​

📍 Location: West Exhibit Hall #17
⏰ Time: 04:30 PM - 06:30 PM

[Yihang Luo @onelineluo]

For more information, check out: https://t.co/3wCrPPGggG

[Yihang Luo @onelineluo]

An exciting week in Hamburg, Germany! Proud to announce that our NTU team emerged victorious, winning the 1st place at the 2023 ISC International Student Virtual Cluster Competition.

#ISC23 #ISC23_SCC #NTU

[Yihang Luo @onelineluo]

P/s: Thank you for EPCC cup, our one is on the way to Singapore.