Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

Added to collectionCV

CC 0

Source arXiv

Computer Vision

Multi-modal Vision Transformers for Object Detection

Multi-modal Transformers Excel at Class-agnostic Object Detection

We advocate that existing methods lack a top-down supervision signal governed by human-understandable semantics.

To bridge this gap, we explore recent multi-modal vision transformers (mvit) that have been trained with aligned image-text pairs.

Our extensive experiments across various domains and novel objects show the state-of-the-art performance of mvits to localize generic objects in images.

Based on these findings, we develop an efficient and flexible mvit architecture using multi-scale feature processing and deformable self-attention that can adaptively generate proposals given a specific language query.

We show the significance of mvit proposals in a diverse range of applications including open-world object detection, salient and camouflageobject detection, supervised and self-supervised detection tasks.

Authors

Muhammad Maaz, Hanoona Rasheed, Salman Khan, Fahad Shahbaz Khan, Rao Muhammad Anwer, Ming-Hsuan Yang

Related Topics

MVIT

Mvits

Learning-free and learning-based approaches

Multi-scale feature processing

Mvit architecture

Multi-modal vision transformers

Read the Paper

Read Paper

Trending

Bookmarks

Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

Added to collectionCV

CC 0

Source arXiv

Computer Vision

Multi-modal Vision Transformers for Object Detection

Multi-modal Transformers Excel at Class-agnostic Object Detection

We advocate that existing methods lack a top-down supervision signal governed by human-understandable semantics.

To bridge this gap, we explore recent multi-modal vision transformers (mvit) that have been trained with aligned image-text pairs.

Our extensive experiments across various domains and novel objects show the state-of-the-art performance of mvits to localize generic objects in images.

Authors

Muhammad Maaz, Hanoona Rasheed, Salman Khan, Fahad Shahbaz Khan, Rao Muhammad Anwer, Ming-Hsuan Yang

◐ Recommended Follows

Qz5210735376

Lynne Zhang

Frederic S

◐ Recommended Follows

◐ Latest Activity