paper-with-me

홈 › Papers

Hands-on Guidance for Distilling Object Detectors

2021-03-26 · Yangyang Qin, Hefei Ling, Zhenghai He, Yuxuan Shi, Lei Wu

Knowledge distillation can lead to deploy-friendly networks against the plagued computational complexity problem, but previous methods neglect the feature hierarchy in detectors. Motivated by this, we propose a general framework for detection distillation. Our method, called Hands-on Guidance Distillation, distills the latent knowledge of all stage features for imposing more comprehensive supervision, and focuses on the essence simultaneously for promoting more intense knowledge absorption. Specifically, a series of novel mechanisms are designed elaborately, including correspondence establishment for consistency, hands-on imitation loss measure and re-weighted optimization from both micro and macro perspectives. We conduct extensive evaluations with different distillation configurations over VOC and COCO datasets, which show better performance on accuracy and speed trade-offs. Meanwhile, feasibility experiments on different structural networks further prove the robustness of our HGD.

📄 PDF Abstract BibTeX arXiv:2103.14337

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationObject

Similar Papers 제목 키워드 기반

Distilling Object Detectors With Global Knowledge

2022-10-17 · Sanli Tang, Zhongyu Zhang, Zhanzhan Cheng, Jing Lu 외

Knowledge distillation learns a lightweight student model that mimics a cumbersome teacher. Existing methods regard the knowledge as the feature of each instance or their relations, which is the instance-level knowledge …

Knowledge DistillationObjectobject-detectionObject Detection

Distilling Object Detectors with Fine-grained Feature Imitation

2019-06-09 · CVPR 2019 6 · Tao Wang, Li Yuan, Xiaopeng Zhang, Jiashi Feng

State-of-the-art CNN based recognition models are often computationally prohibitive to deploy on low-end devices. A promising high level approach tackling this limitation is knowledge distillation, which let small studen…

Knowledge DistillationObjectobject-detectionObject Detection

Distilling Object Detectors via Decoupled Features

2021-03-26 · CVPR 2021 1 · Jianyuan Guo, Kai Han, Yunhe Wang, Han Wu 외

Knowledge distillation is a widely used paradigm for inheriting information from a complicated teacher network to a compact student network and maintaining the strong performance. Different from image classification, obj…

image-classificationImage ClassificationKnowledge DistillationObject+2

AnyDexRT: Calibration-Free Dexterous Hand Retargeting with Few-Shot Human Guidance

2026-07-09 · Chenxi Wang, Ying Feng, Hongjie Fang, Shangning Xia 외 arxiv

Teleoperation is a key interface for controlling dexterous robotic hands and collecting demonstrations for imitation learning. Its effectiveness largely depends on kinematic retargeting, which maps operator hand motions …

RHanDS: Refining Malformed Hands for Generated Images with Decoupled Structure and Style Guidance

2024-04-22 · Chengrui Wang, PengFei Liu, Min Zhou, Ming Zeng 외

Although diffusion models can generate high-quality human images, their applications are limited by the instability in generating hands with correct structures. Some previous works mitigate the problem by considering han…