Hierarchical Context Transformer for Multi-level Semantic Scene Understanding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Hao, Luoying, Hu, Yan, Yue, Yang, Wu, Li, Fu, Huazhu, Duan, Jinming, Liu, Jiang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910837989441536
author Hao, Luoying
Hu, Yan
Yue, Yang
Wu, Li
Fu, Huazhu
Duan, Jinming
Liu, Jiang
author_facet Hao, Luoying
Hu, Yan
Yue, Yang
Wu, Li
Fu, Huazhu
Duan, Jinming
Liu, Jiang
contents A comprehensive and explicit understanding of surgical scenes plays a vital role in developing context-aware computer-assisted systems in the operating theatre. However, few works provide systematical analysis to enable hierarchical surgical scene understanding. In this work, we propose to represent the tasks set [phase recognition --> step recognition --> action and instrument detection] as multi-level semantic scene understanding (MSSU). For this target, we propose a novel hierarchical context transformer (HCT) network and thoroughly explore the relations across the different level tasks. Specifically, a hierarchical relation aggregation module (HRAM) is designed to concurrently relate entries inside multi-level interaction information and then augment task-specific features. To further boost the representation learning of the different tasks, inter-task contrastive learning (ICL) is presented to guide the model to learn task-wise features via absorbing complementary information from other tasks. Furthermore, considering the computational costs of the transformer, we propose HCT+ to integrate the spatial and temporal adapter to access competitive performance on substantially fewer tunable parameters. Extensive experiments on our cataract dataset and a publicly available endoscopic PSI-AVA dataset demonstrate the outstanding performance of our method, consistently exceeding the state-of-the-art methods by a large margin. The code is available at https://github.com/Aurora-hao/HCT.
format Preprint
id arxiv_https___arxiv_org_abs_2502_15184
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Context Transformer for Multi-level Semantic Scene Understanding
Hao, Luoying
Hu, Yan
Yue, Yang
Wu, Li
Fu, Huazhu
Duan, Jinming
Liu, Jiang
Computer Vision and Pattern Recognition
A comprehensive and explicit understanding of surgical scenes plays a vital role in developing context-aware computer-assisted systems in the operating theatre. However, few works provide systematical analysis to enable hierarchical surgical scene understanding. In this work, we propose to represent the tasks set [phase recognition --> step recognition --> action and instrument detection] as multi-level semantic scene understanding (MSSU). For this target, we propose a novel hierarchical context transformer (HCT) network and thoroughly explore the relations across the different level tasks. Specifically, a hierarchical relation aggregation module (HRAM) is designed to concurrently relate entries inside multi-level interaction information and then augment task-specific features. To further boost the representation learning of the different tasks, inter-task contrastive learning (ICL) is presented to guide the model to learn task-wise features via absorbing complementary information from other tasks. Furthermore, considering the computational costs of the transformer, we propose HCT+ to integrate the spatial and temporal adapter to access competitive performance on substantially fewer tunable parameters. Extensive experiments on our cataract dataset and a publicly available endoscopic PSI-AVA dataset demonstrate the outstanding performance of our method, consistently exceeding the state-of-the-art methods by a large margin. The code is available at https://github.com/Aurora-hao/HCT.
title Hierarchical Context Transformer for Multi-level Semantic Scene Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.15184