Papers
arxiv:2412.10360

Apollo: An Exploration of Video Understanding in Large Multimodal Models

Published on Dec 13
· Submitted by orrzohar on Dec 16
#1 Paper of the day
Authors:
,
,
,
,

Abstract

Despite the rapid integration of video perception capabilities into Large Multimodal Models (LMMs), the underlying mechanisms driving their video understanding remain poorly understood. Consequently, many design decisions in this domain are made without proper justification or analysis. The high computational cost of training and evaluating such models, coupled with limited open research, hinders the development of video-LMMs. To address this, we present a comprehensive study that helps uncover what effectively drives video understanding in LMMs. We begin by critically examining the primary contributors to the high computational requirements associated with video-LMM research and discover Scaling Consistency, wherein design and training decisions made on smaller models and datasets (up to a critical size) effectively transfer to larger models. Leveraging these insights, we explored many video-specific aspects of video-LMMs, including video sampling, architectures, data composition, training schedules, and more. For example, we demonstrated that fps sampling during training is vastly preferable to uniform frame sampling and which vision encoders are the best for video representation. Guided by these findings, we introduce Apollo, a state-of-the-art family of LMMs that achieve superior performance across different model sizes. Our models can perceive hour-long videos efficiently, with Apollo-3B outperforming most existing 7B models with an impressive 55.1 on LongVideoBench. Apollo-7B is state-of-the-art compared to 7B LMMs with a 70.9 on MLVU, and 63.3 on Video-MME.

Community

Paper author Paper submitter
This comment has been hidden
Paper author Paper submitter
This comment has been hidden

png-clipart-racing-bicycle-bicycle-parking-rack-cycling-bikes-bicycle-frame-bicycle-thumbnail.png

Cosa si vede nella foto

can i know when the code and model weights will be available again?

deleted
This comment has been hidden

Unfortunately, the code, model and space demo have all been pulled for some reason (PR of meta releasing a Qwen based model?).

·

The weights have been mirrored to https://huggingface.co/GoodiesHere and the inference code has been mirrored to https://huggingface.co/manysuch-cases/Apollo-Github-Files.

Sign up or log in to comment

Models citing this paper 4

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2412.10360 in a dataset README.md to link it from this page.

Spaces citing this paper 1

Collections including this paper 28