Data

Tools

Indexes

Applications

Experiments

Open Source Science

lda-aqu

[MM2024] LDA-AQU: Adaptive Query-guided Upsampling via Local Deformable Attention

https://github.com/duzw9311/lda-aqu

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (8.8%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

[MM2024] LDA-AQU: Adaptive Query-guided Upsampling via Local Deformable Attention

Basic Info
  • Host: GitHub
  • Owner: duzw9311
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 7.45 MB
Statistics
  • Stars: 10
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed about 1 year ago
Metadata Files
Readme License Citation

README.md

LDA-AQU: Adaptive Query-guided Upsampling via Local Deformable Attention

This repository presents the official PyTorch implementation of LDA-AQU (MM'2024).

In this paper, we propose LDA-AQU, which incorporates local self-attention into the feature upsampling process and introduces local deformation capabilities to mitigate the semantic gap between interpolation points and their neighboring points selected during feature reassembly.

Here is the performance comparison of various upsampling operators integrated into the Faster RCNN detector on the MS COCO dataset.

Here is the overall architecture of the proposed LDA-AQU.

Installation

Please see get_started.md for the basic usage of MMDetection.

Training

bash bash tools/dist_train.sh configs/lda_aqu/fasterrcnn_r50_lau.py 4

Testing

bash python tools/test.py configs/lda_aqu/fasterrcnn_r50_lau.py work_dirs/lda_aqu/latest.pth --eval bbox

Weight

Model | AP | Link1 | Link2 | --- |:---:|:---:|:---: fasterrcnnr50lau | 39.2 | BaiduNetDisk | GoogleDrive

Acknowledgement

This repository is built upon the MMDetection library.

Citation

If you find this paper helpful for your project, we'd appreciate it if you could cite it. @inproceedings{du2024lda, title={LDA-AQU: Adaptive Query-guided Upsampling via Local Deformable Attention}, author={Du, Zewen and Hu, Zhenjiang and Zhao, Guiyu and Jin, Ying and Ma, Hongbin}, booktitle={Proceedings of the 32nd ACM International Conference on Multimedia}, pages={4919--4927}, year={2024} }

Owner

  • Name: duzw1
  • Login: duzw9311
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
  - name: "MMDetection Contributors"
title: "OpenMMLab Detection Toolbox and Benchmark"
date-released: 2018-08-22
url: "https://github.com/open-mmlab/mmdetection"
license: Apache-2.0

GitHub Events

Total
  • Issues event: 1
  • Watch event: 10
  • Push event: 1
Last Year
  • Issues event: 1
  • Watch event: 10
  • Push event: 1