LEDGER Builds Persistent 3D Object Memory from Egocentric Video

Johns Hopkins researchers introduced LEDGER, which builds persistent 3D object memory from egocentric video, answering questions without replaying footage. Described in the paper "Never Look Back," it combines long-horizon egocentric descriptions, geometry, and event records.

2026-10-08 ~ 2026-10-08 · 2 related posts

• LEDGER builds persistent 3D object memory from egocentric video, answers questions without rewatching — mangahomanga · 2026-10-08

1 near-duplicate retellings: mangahomanga