ScaleKD: Distilling Scale-Aware Knowledge in Small Object Detector
Despite the prominent success of general object detection, the performance and efficiency of Small Object Detection (SOD) are still unsatisfactory. Unlike existing works that struggle to balance the trade-off between inference speed and SOD performance, in this paper, we propose a novel Scale-aware Knowledge Distillation (ScaleKD), which transfers knowledge of a complex teacher model to a compact student model. We design two novel modules to boost the quality of knowledge transfer in distillation for SOD: 1) a scale-decoupled feature distillation module that disentangled teacher's feature representation into multi-scale embedding that enables explicit feature mimicking of the student model on small objects. 2) a cross-scale assistant to refine the noisy and uninformative bounding boxes prediction student models, which can mislead the student model and impair the efficacy of knowledge distillation. A multi-scale cross-attention layer is established to capture the multi-scale semantic information to improve the student model. We conduct experiments on COCO and VisDrone datasets with diverse types of models, i.e., two-stage and one-stage detectors, to evaluate our proposed method. Our ScaleKD achieves superior performance on general detection performance and obtains spectacular improvement regarding the SOD performance.
Code (0)
등록된 구현이 없습니다.
Tasks
Knowledge Distillationobject-detectionObject DetectionSmall Object DetectionTransfer LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
ScaleKD: Strong Vision Transformers Could Be Excellent Teachers
In this paper, we question if well pre-trained vision transformer (ViT) models could be used as teachers that exhibit scalable properties to advance cross architecture knowledge distillation (KD) research, in the context…
image-classificationImage ClassificationKnowledge DistillationDistilling Word Embeddings: An Encoding Approach
Distilling knowledge from a well-trained cumbersome network to a small one has recently become a new research topic, as lightweight neural networks with high performance are particularly in need in various resource-restr…
Word EmbeddingsDistill, Adapt, Distill: Training Small, In-Domain Models for Neural Machine Translation
We explore best practices for training small, memory efficient machine translation models with sequence-level knowledge distillation in the domain adaptation setting. While both domain adaptation and knowledge distillati…
Domain AdaptationKnowledge DistillationMachine TranslationTranslationGraph-based Knowledge Distillation by Multi-head Attention Network
Knowledge distillation (KD) is a technique to derive optimal performance from a small student network (SN) by distilling knowledge of a large teacher network (TN) and transferring the distilled knowledge to the small SN.…
Inductive BiasKnowledge DistillationMulti-Task LearningTransfer LearningMind's Mirror: Distilling Self-Evaluation Capability and Comprehensive Thinking from Large Language Models
Large language models (LLMs) have achieved remarkable advancements in natural language processing. However, the massive scale and computational demands of these models present formidable challenges when considering their…
Transfer Learning