Loading video...

Video Failed to Load

Go Home

NEVER LOOK BACK! Introducing LEDGER: persistent 3D object memory from egocentric videos. Remember where things were, when they moved, and what they held -> then answer questions without looking back at the video!

12,620 views • 17 days ago •via X (Twitter)

8 Comments

Homanga Bharadhwaj's profile picture
Homanga Bharadhwaj17 days ago

This was a fun project led by @Shra_1_C along with William Paul, and Johns Hopkins colleagues @suchisaria and Rama Chellappa Thread with details from @Shra_1_C below

Dominik's profile picture
Dominik17 days ago

object permanence for agents is so underrated. does ledger track things that leave the frame for hours, like keys that got moved by someone else while the wearer was out?

Sani Ai Tech's profile picture
Sani Ai Tech17 days ago

Persistent 3D memory could be a huge step for egocentric AI systems

Vera Ai | Tools & Updates's profile picture
Vera Ai | Tools & Updates16 days ago

Persistent 3D memory could be a major leap for embodied AI reasoning

LM10/7's profile picture
LM10/716 days ago

Actually solid, also real-time implementation will make this🤌🏼

Zara Tech & Tools's profile picture
Zara Tech & Tools16 days ago

Smart approach to remembering object history

Vantix AI Agency's profile picture
Vantix AI Agency16 days ago

Persistent 3D object memory from egocentric video is wild answering where things were without rewatching is huge

Veyra | AI & Tech's profile picture
Veyra | AI & Tech16 days ago

Egocentric video memory that actually persists

Related Videos