Two head structures (ie fully connected head and convolution head) have been widely used in R-CNN based detectors for classification and localization tasks. However, there is a lack …
Z Zheng, R Ye, P Wang, D Ren… - Proceedings of the …, 2022 - openaccess.thecvf.com
Abstract Knowledge distillation (KD) has witnessed its powerful capability in learning compact models in object detection. Previous KD methods for object detection mostly focus …
The field of object detection has made great progress in recent years. Most of these improvements are derived from using a more sophisticated convolutional neural network …
J Ma, Y Ushiku, M Sagara - … of the IEEE/CVF Conference on …, 2022 - openaccess.thecvf.com
In this study, we partially reannotate conventional benchmark datasets for object detection and check whether there is performance improvement/drop compared with the original …
J Nie, RM Anwer, H Cholakkal… - Proceedings of the …, 2019 - openaccess.thecvf.com
We propose a single-stage detection framework that jointly tackles the problem of multi- scale object detection and class imbalance. Rather than designing deeper networks, we …
M Ye, L Ke, S Li, YW Tai, CK Tang… - Proceedings of the …, 2023 - openaccess.thecvf.com
Object localization in general environments is a fundamental part of vision systems. While dominating on the COCO benchmark, recent Transformer-based detection methods are not …
Z Yang, S Liu, H Hu, L Wang… - Proceedings of the IEEE …, 2019 - openaccess.thecvf.com
Modern object detectors rely heavily on rectangular bounding boxes, such as anchors, proposals and the final predictions, to represent objects at various recognition stages. The …
S Li, C He, R Li, L Zhang - … of the IEEE/CVF conference on …, 2022 - openaccess.thecvf.com
Label assignment (LA), which aims to assign each training sample a positive (pos) and a negative (neg) loss weight, plays an important role in object detection. Existing LA methods …
H Li, Z Wu, C Zhu, C Xiong… - Proceedings of the …, 2020 - openaccess.thecvf.com
State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their …