@HuggingPapers
DailyPapers
4 days
EgoX: Generate immersive first-person video from any third-person clip A novel framework from KAIST AI & Seoul National University that leverages video diffusion models to transform a single exocentric video into a realistic egocentric view. See it in action!
15
182
1K

Replies

@HuggingPapers
DailyPapers
4 days
EgoX achieves coherent & realistic egocentric video generation, proving robust across diverse, unseen scenarios! Explore how this framework overcomes extreme camera pose challenges for immersive experiences:
0
11
54
@heyellieday
ELLIE X
2 days
1
0
1
@DarwinexZero
Darwinex Zero
7 months
Any reason not to secure a €100k allocation permanently? 🤔 👉 If you're an independent trader committed to growth, try our platform. Build something that lasts! Start at
7
7
101
@zsakib_
Sakib
3 days
@HuggingPapers Woahhhh
0
0
1
@DrTunnels
Andy Cheng
3 days
@HuggingPapers Am I the only person thinking "that" application?
2
0
14
@jolfss
Sean Brynjólfsson
3 days
@HuggingPapers Love love love it
0
0
0
@codewithimanshu
Himanshu Kumar
3 days
@HuggingPapers First-person video generation from third-person clips achieves high realism.
0
0
1
@14a0x
14aØ
3 days
@HuggingPapers That table tennis view deserves a gold medal.
0
0
0
@TremereProblem
𓋹 𝚁𝚑𝚞𝚊𝚗 𝙿𝚊𝚐𝚊𝚗𝚘 𓋹
3 days
@HuggingPapers This is wild
0
0
0
@NigelHiggs7
Nigel Higgs
3 days
@HuggingPapers Were the hell do you even get the training data for this? Lol
2
0
14
@scavxxx
scavx
3 days
@HuggingPapers This is lowkey brilliant
0
0
0
@suhrabautomates
Suhrab Khan⚡️
3 days
@HuggingPapers Incredible innovation! EgoX opens up new possibilities for immersive video experiences by turning standard clips into realistic first-person perspectives.
0
0
0
@aries_logistics
Aries Worldwide Logistics
3 months
Aries is transforming logistics with modern tracking and unmatched client experience. Get a Quote or Book Now.
0
18
163
@CoeltheSecond
earendil
3 days
@HuggingPapers amazing
0
0
0
@victormustar
Victor M
2 days
Microsoft TRELLIS.2 is here 🔥 • Single image → textured 3D mesh • 4B params, flow-matching transformer • Up to 1536³ resolution • Open weights, MIT licensed ⬇️ Demo available on Hugging Face
63
400
3K
@measure_plan
AA
11 hours
the secret life of boids using gemini 3 flash to simulate simple rules that lead to emergent behaviours (attraction, repulsion, cohesion)
31
72
752
@AIatMeta
AI at Meta
13 hours
We’re open-sourcing Perception Encoder Audiovisual (PE-AV), the technical engine that helps drive SAM Audio’s state-of-the-art audio separation. Built on our Perception Encoder model from earlier this year, PE-AV integrates audio with visual perception, achieving
41
130
1K
@calebarclay
Caleb Barclay
2 days
Announcing Arcway ⊹ ࣪ ˖ A real-time 3D engine where anyone can design a home. It is a simulated 3D world where buyers explore, change, and decide as light, physics, building rules, and real products move as one.
64
127
2K