DeepAI AI Chat
Log In Sign Up

Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval

by   Max Bain, et al.

Our objective in this work is video-text retrieval - in particular a joint embedding that enables efficient text-to-video retrieval. The challenges in this area include the design of the visual architecture and the nature of the training data, in that the available large scale video-text training datasets, such as HowTo100M, are noisy and hence competitive performance is achieved only at scale through large amounts of compute. We address both these challenges in this paper. We propose an end-to-end trainable model that is designed to take advantage of both large-scale image and video captioning datasets. Our model is an adaptation and extension of the recent ViT and Timesformer architectures, and consists of attention in both space and time. The model is flexible and can be trained on both image and video text datasets, either independently or in conjunction. It is trained with a curriculum learning schedule that begins by treating images as 'frozen' snapshots of video, and then gradually learns to attend to increasing temporal context when trained on video datasets. We also provide a new video-text pretraining dataset WebVid-2M, comprised of over two million videos with weak captions scraped from the internet. Despite training on datasets that are an order of magnitude smaller, we show that this approach yields state-of-the-art results on standard downstream video-retrieval benchmarks including MSR-VTT, MSVD, DiDeMo and LSMDC.


page 5

page 14


Learning Audio-Video Modalities from Image Captions

A major challenge in text-video and text-audio retrieval is the lack of ...

A Straightforward Framework For Video Retrieval Using CLIP

Video Retrieval is a challenging task where a text query is matched to a...

CNN-VWII: An Efficient Approach for Large-Scale Video Retrieval by Image Queries

This paper aims to solve the problem of large-scale video retrieval by a...

Learning a Text-Video Embedding from Incomplete and Heterogeneous Data

Joint understanding of video and language is an active research area wit...

Rethinking Video ViTs: Sparse Video Tubes for Joint Image and Video Learning

We present a simple approach which can turn a ViT encoder into an effici...

MDMMT: Multidomain Multimodal Transformer for Video Retrieval

We present a new state-of-the-art on the text to video retrieval task on...

Fighting FIRe with FIRE: Assessing the Validity of Text-to-Video Retrieval Benchmarks

Searching vast troves of videos with textual descriptions is a core mult...

Code Repositories


Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval [ICCV'21]

view repo


Large-scale text-video dataset.

view repo