Hai Pham

MVA Master @ ENS. Ex-Intern @ Qualcomm AI Research. Working on 3D computer vision and generative models.

[Hai Pham @haiphamcse]

@trungdt880 amazing work!

[Hai Pham @haiphamcse]

@sharifshameem Hi can you share the blender file : ) It's great seeing the renderings but im curious ab the true underlying 3D

[Hai Pham @haiphamcse]

@mweinbach Hi can you share the blender file? Tks

[Hai Pham @haiphamcse]

@bilawalsidhu nice! pipeline similar to colmap or how does it work

[Hai Pham @haiphamcse]

@MadanKhd nice. Which app did you use to write the blog? Looks very elegant

[ELLIS @ELLISforEurope]

🎓 We're thrilled to welcome 121 new PhD students to the ELLIS network from our 2025/2026 call — selected from a record-breaking 2,100+ submitted applications, a 43% increase from last year.

👉 Read the full article for more details: https://t.co/II3rkmOIOm

[Hai Pham @haiphamcse]

@sbh_studio Nice capture! Which model/framework did you use for this?

[Angela Dai @angelaqdai]

Come by our ScanNet++ workshop @CVPR June 3 in 710, 1:00pm onwards!

5 exciting keynotes on world models, NVS, 3D gen, understanding and more, from @taiyasaki @davnov134 @orlitany @PeterHedman3 @RamananDeva and talks from NVS+semantics benchmark winners:
https://t.co/DtU3Qg3jqd

[Hai Pham @haiphamcse]

@HenriqueGoisDev Hi, great first try. How did you parameterize the mesh? Directly adding noise to edges and vertices?

[Hai Pham @haiphamcse]

Super impressive reconstruction work from DeepMind 👀
D4RT isn’t just accurate — the sharp, high-res details really stand out.
Also great to see the team using our proposed depth sharpness metric (https://t.co/HRHejccH8k)😁

[Chuhan Zhang @ChuhanZhang5]

A SINGLE encoder + decoder for all the 4D tasks!
We release 🎯 D4RT (Dynamic 4D Reconstruction and Tracking).

📍 A simple, unified interface for 3D tracking, depth, and pose
🌟 SOTA results on 4D reconstruction & tracking
🚀 Up to 100x faster pose estimation than prior works

[Hai Pham @haiphamcse]

@artonson Hi, is it possible for other EU students to apply for the pos? Tks!

[Hai Pham @haiphamcse]

@arkitus have faith (and a lot of gpus)

[Hai Pham @haiphamcse]

Our paper's code is finally released (under the shiny Qualcomm github)! Feel free to sharpen your metric depth estimation

Code: https://t.co/HRHejccH8k

[Phong Nguyen-Ha @PhongStormVN]

Can we achieve both accurate depth maps and still maintain sharp edges ? I am proud to present our latest work SharpDepth, a novel monocular depth estimation approach that combines the metric accuracy of discriminative methods with the boundary sharpness of generative models.

[Voxel51 @Voxel51]

One of the biggest bottlenecks in deploying visual AI and computer vision is annotation, which can be both costly and time-consuming. Today, we’re introducing Verified Auto Labeling, a new approach to AI-assisted annotation that achieves up to 95% of human-level performance while cutting labeling costs by up to 100,000x and time by 5,000x.

Read the full paper: https://t.co/eKc1sALnV3

[Hai Pham @haiphamcse]

Our team at Qualcomm AI Research Vietnam is at Nashville for @CVPR🤠
I'll be presenting SharpDepth on 2 occasions:
- 3:30pm @ June 12th: CVEU Workshop, poster #211 in Hall D
- 5:00pm @ June 14th: Main conf, poster #119 in Hall D
Feel free to come & chat! I'm also down to grab a☕

[Phong Nguyen-Ha @PhongStormVN]

Can we achieve both accurate depth maps and still maintain sharp edges ? I am proud to present our latest work SharpDepth, a novel monocular depth estimation approach that combines the metric accuracy of discriminative methods with the boundary sharpness of generative models.

[#CVPR2026 @CVPR]

Countdown’s on: Less than a month until the BIGGEST #computervision event of the year—#CVPR2025 hits Nashville!

[Hai Pham @haiphamcse]

@CVPR Hi there, firstly thanks the organizing committee for the effort! I want to change my poster presentation schedule to an earlier session (due to us not being able to book a hotel to stay at Nashville on the presenting date), who can I email to change this? Tks!

[Hai Pham @haiphamcse]

@gulvarol Hi, the program looks very interesting! I was wondering about admission chances for someone with a 240 ECTS Bachelor of Engineering and top-tier publications (AAAI, CVPR). I saw that an M1 is formally required—can exceptional candidates without one still be considered?

[Hai Pham @haiphamcse]

@CVPR Thanks for the great initiative! When can we expect to hear results?

[Hai Pham @haiphamcse]

@AdamKortylewski @CVPR Thanks for replying. So just to verify. For accepted CVPR papers, the flow is to convert it into a 4 page paper -> submit to openreview right?