Towards Real-Time Open-Vocabulary Video Instance Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yan, Bin, Sundermeyer, Martin, Tan, David Joseph, Lu, Huchuan, Tombari, Federico
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913599280119808
author Yan, Bin
Sundermeyer, Martin
Tan, David Joseph
Lu, Huchuan
Tombari, Federico
author_facet Yan, Bin
Sundermeyer, Martin
Tan, David Joseph
Lu, Huchuan
Tombari, Federico
contents In this paper, we address the challenge of performing open-vocabulary video instance segmentation (OV-VIS) in real-time. We analyze the computational bottlenecks of state-of-the-art foundation models that performs OV-VIS, and propose a new method, TROY-VIS, that significantly improves processing speed while maintaining high accuracy. We introduce three key techniques: (1) Decoupled Attention Feature Enhancer to speed up information interaction between different modalities and scales; (2) Flash Embedding Memory for obtaining fast text embeddings of object categories; and, (3) Kernel Interpolation for exploiting the temporal continuity in videos. Our experiments demonstrate that TROY-VIS achieves the best trade-off between accuracy and speed on two large-scale OV-VIS benchmarks, BURST and LV-VIS, running 20x faster than GLEE-Lite (25 FPS v.s. 1.25 FPS) with comparable or even better accuracy. These results demonstrate TROY-VIS's potential for real-time applications in dynamic environments such as mobile robotics and augmented reality. Code and model will be released at https://github.com/google-research/troyvis.
format Preprint
id arxiv_https___arxiv_org_abs_2412_04434
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards Real-Time Open-Vocabulary Video Instance Segmentation
Yan, Bin
Sundermeyer, Martin
Tan, David Joseph
Lu, Huchuan
Tombari, Federico
Computer Vision and Pattern Recognition
In this paper, we address the challenge of performing open-vocabulary video instance segmentation (OV-VIS) in real-time. We analyze the computational bottlenecks of state-of-the-art foundation models that performs OV-VIS, and propose a new method, TROY-VIS, that significantly improves processing speed while maintaining high accuracy. We introduce three key techniques: (1) Decoupled Attention Feature Enhancer to speed up information interaction between different modalities and scales; (2) Flash Embedding Memory for obtaining fast text embeddings of object categories; and, (3) Kernel Interpolation for exploiting the temporal continuity in videos. Our experiments demonstrate that TROY-VIS achieves the best trade-off between accuracy and speed on two large-scale OV-VIS benchmarks, BURST and LV-VIS, running 20x faster than GLEE-Lite (25 FPS v.s. 1.25 FPS) with comparable or even better accuracy. These results demonstrate TROY-VIS's potential for real-time applications in dynamic environments such as mobile robotics and augmented reality. Code and model will be released at https://github.com/google-research/troyvis.
title Towards Real-Time Open-Vocabulary Video Instance Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.04434