agentic-learning-ai-lab/lifelong-memory
Code for LifelongMemory: Leveraging LLMs for Answering Queries in Long-form Egocentric Videos
This tool helps you quickly find specific moments and answer questions about actions captured in long, first-person video recordings. You provide your egocentric video footage (like from a bodycam) and your questions in natural language, and it outputs precise answers or timestamps for relevant events. It's designed for researchers or analysts who need to efficiently review extensive subjective video data.
Use this if you need to extract specific information or answer questions from many hours of first-person video content.
Not ideal if your videos are not egocentric (first-person perspective) or if you need to process short-form, general video content.
Stars
28
Forks
—
Language
Python
License
MIT
Category
Last pushed
Oct 27, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/agentic-learning-ai-lab/lifelong-memory"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
KimMeen/Time-LLM
[ICLR 2024] Official implementation of " 🦙 Time-LLM: Time Series Forecasting by Reprogramming...
om-ai-lab/VLM-R1
Solve Visual Understanding with Reinforced VLMs
bytedance/SALMONN
SALMONN family: A suite of advanced multi-modal LLMs
NVlabs/OmniVinci
OmniVinci is an omni-modal LLM for joint understanding of vision, audio, and language.
fixie-ai/ultravox
A fast multimodal LLM for real-time voice