paper-with-me

홈 › Papers

Architectural Insights into Knowledge Distillation for Object Detection: A Comprehensive Review

2025-08-05 · Mahdi Golizadeh, Nassibeh Golizadeh, Mohammad Ali Keyvanrad, Hossein Shirazi arxiv

Object detection has achieved remarkable accuracy through deep learning, yet these improvements often come with increased computational cost, limiting deployment on resource-constrained devices. Knowledge Distillation (KD) provides an effective solution by enabling compact student models to learn from larger teacher models. However, adapting KD to object detection poses unique challenges due to its dual objectives-classification and localization-as well as foreground-background imbalance and multi-scale feature representation. This review introduces a novel architecture-centric taxonomy for KD methods, distinguishing between CNN-based detectors (covering backbone-level, neck-level, head-level, and RPN/RoI-level distillation) and Transformer-based detectors (including query-level, feature-level, and logit-level distillation). We further evaluate representative methods using the MS COCO and PASCAL VOC datasets with mAP@0.5 as performance metric, providing a comparative analysis of their effectiveness. The proposed taxonomy and analysis aim to clarify the evolving landscape of KD in object detection, highlight current challenges, and guide future research toward efficient and scalable detection systems.

📄 PDF Abstract BibTeX arXiv:2508.03317

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationObject Detection

Similar Papers 제목 키워드 기반

Empirical Evaluation of Knowledge Distillation from Transformers to Subquadratic Language Models

2025-04-19 · Patrick Haller, Jonas Golde, Alan Akbik

Knowledge distillation is a widely used technique for compressing large language models (LLMs) by training a smaller student model to mimic a larger teacher model. Typically, both the teacher and student are Transformer-…

Knowledge DistillationState Space ModelsTransfer Learning

Geometric Limits of Knowledge Distillation: A Minimum-Width Theorem via Superposition Theory

2026-04-05 · Nilesh Sarkar, Dawar Jyoti Deka arxiv

Knowledge distillation compresses large teachers into smaller students, but performance saturates at a loss floor that persists across training methods and objectives. We argue this floor is geometric: neural networks re…

Knowledge Distillation

Towards Optimal Trade-offs in Knowledge Distillation for CNNs and Vision Transformers at the Edge

2024-06-25 · John Violos, Symeon Papadopoulos, Ioannis Kompatsiaris

This paper discusses four facets of the Knowledge Distillation (KD) process for Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) architectures, particularly when executed on edge devices with constrained…

Knowledge Distillation

Scalpel-SAM: A Semi-Supervised Paradigm for Adapting SAM to Infrared Small Object Detection

2025-12-27 · Zihan Liu, Xiangning Ren, Dezhang Kong, Yipeng Zhang 외 arxiv

Infrared small object detection urgently requires semi-supervised paradigms due to the high cost of annotation. However, existing methods like SAM face significant challenges of domain gaps, inability of encoding physica…

Knowledge DistillationSmall Object Detection

Towards a Unified View of Affinity-Based Knowledge Distillation

2022-09-30 · Vladimir Li, Atsuto Maki

Knowledge transfer between artificial neural networks has become an important topic in deep learning. Among the open questions are what kind of knowledge needs to be preserved for the transfer, and how it can be effectiv…

image-classificationImage ClassificationKnowledge DistillationRelation+1