Deep Detector

Deep detectors, primarily convolutional neural networks, are used for object detection and segmentation in various applications, including malware analysis, underwater imaging, and autonomous driving. Current research focuses on improving their robustness against adversarial attacks (e.g., through randomized smoothing or adversarial training) and addressing limitations in handling imbalanced datasets (e.g., via modified loss functions). These advancements are crucial for enhancing the reliability and accuracy of deep detectors in real-world scenarios, impacting fields ranging from cybersecurity to robotics and medical imaging.

Papers

August 17, 2023

Towards a Practical Defense against Adversarial Attacks on Deep Learning-based Malware Detectors via Randomized Smoothing
Daniel Gibert, Giulio Zizzo, Quan Le
Adversarial Attack Malware Detection Randomized Smoothing Adversarial Malware Based Malware Deep Detector

July 7, 2023

Joint Perceptual Learning for Enhancement and Object Detection in Underwater Scenarios
Chenping Fu, Wanqi Yuan, Jiewen Xiao, Risheng Liu, Xin Fan
Feature Enhancement Underwater Environment Joint Learning Underwater Object Detection Underwater Detection Deep Detector

February 19, 2023

Accelerated Video Annotation driven by Deep Detector and Tracker
Eric Price, Aamir Ahmad
Object Detection Video Annotation Ground Truth Annotation External Tracker Object Tracker Deep Detector

July 22, 2022

Long-tailed Instance Segmentation using Gumbel Optimized Loss
Konstantinos Panagiotis Alexandridis, Jiankang Deng, Anh Nguyen, Shan Luo
Long Tailed Deep Detector

May 5, 2022

Evaluating Context for Deep Object Detectors
Osman Semih Kayhan, Jan C. van Gemert
Object Detector Deep Learning Based Object Two Stage Object Deep Detector

February 10, 2022

Adversarial Attack and Defense of YOLO Detectors in Autonomous Driving Scenarios
Jung Im Choi, Qing Tian
Adversarial Attack YOLO Object Visual Detection Deep Detector