DIGCN: A Dynamic Interaction Graph Convolutional Network Based on Learnable Proposals for Object Detection

计算机科学 目标检测 对象(语法) 图形 集合(抽象数据类型) 探测器 人工智能 计算复杂性理论 模式识别(心理学) 算法 理论计算机科学 电信 程序设计语言
作者
Pingping Cao,Yanping Zhu,Yuhao Jin,Benkun Ruan,Qiang Niu
出处
期刊:Journal of Artificial Intelligence Research [AI Access Foundation]
卷期号:79: 1091-1112 被引量:3
标识
DOI:10.1613/jair.1.15698
摘要

We propose a Dynamic Interaction Graph Convolutional Network (DIGCN), an image object detection method based on learnable proposals and GCN. Existing object detection methods usually work on dense candidates, resulting in redundant and near-duplicate results. Meanwhile, non-maximum suppression post-processing operations are required to eliminate negative effects, which increases the computational complexity. Although the existing sparse detector avoids cumbersome post-processing operations, it ignores the potential relationship between objects and proposals, which hinders detection accuracy improvement. Therefore, we propose a dynamic interaction GCN module in the DIGCN, which performs dynamic interaction and relational modeling on the proposal boxes and proposal features to improve the object detection accuracy. In addition, we introduce a learnable proposal method with a sparse set of learned object proposals to eliminate a huge number of hand-designed object candidates, avoiding complicated tasks such as object candidate design and many-to-one label assignment, and reducing object detection model complexity to a certain extent. DIGCN demonstrates accuracy and run-time performance on par with the well-established and highly optimized detector baselines on the challenging COCO dataset, e.g. with the ResNet-101FPN as the backbone our method attains the accuracy of 46.5 AP while processing 13 frames per second. Our work provides a new method for object detection research.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
zhao发布了新的文献求助10
1秒前
1秒前
1秒前
迟暮发布了新的文献求助10
1秒前
无情向薇完成签到,获得积分10
1秒前
2秒前
内向如松发布了新的文献求助10
2秒前
2秒前
居里夫人发布了新的文献求助10
2秒前
3秒前
3秒前
彭于晏的应助被dff采纳,获得10
4秒前
HHW完成签到,获得积分10
4秒前
4秒前
4秒前
sci小神童完成签到,获得积分10
4秒前
4秒前
张美超发布了新的文献求助10
4秒前
4秒前
4秒前
灞波儿奔的应助被corleeang采纳,获得10
5秒前
5秒前
5秒前
yue发布了新的文献求助10
5秒前
1111发布了新的文献求助10
5秒前
涂银冰完成签到,获得积分10
5秒前
6秒前
研友_LkY7BZ完成签到,获得积分10
6秒前
哇哈哈完成签到,获得积分10
6秒前
6秒前
Min完成签到,获得积分10
7秒前
所所的应助被微笑采纳,获得10
7秒前
7秒前
NexusExplorer的应助被温暖砖头采纳,获得10
7秒前
7秒前
7秒前
7秒前
7秒前
jiao完成签到,获得积分10
7秒前
黄小宇发布了新的文献求助10
8秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Improving the wear resistance of vehicle gear oil by oil soluble LaF3 nanoparticles and typical additives of gear oil 500
Boundaries and Bodhisattvas 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8183046
求助须知:如何正确求助?哪些是违规求助? 9572370
关于积分的说明 21824411
捐赠科研通 7779694
什么是DOI,文献DOI怎么找? 3433792
关于科研通互助平台的介绍 2558024
邀请新用户注册赠送积分活动 2479971