Abstract:Object detection based on Knowledge Distillation can enhance the capabilities and performance of 5G and 6G networks in various domains, such as autonomous vehicles, smart surveillance, and augmented reality. The integration of object detection with Knowledge Distillation techniques is expected to play a pivotal role in realizing the full potential of these networks. This study presents Shared Knowledge Distillation (Shared-KD) as a solution to overcome optimization challenges caused by disparities in cross-layer features between teacher–student networks. The significant gaps in intermediate-level features between teachers and students present a considerable obstacle to the efficacy of distillation. To tackle this issue, we draw inspiration from collaborative learning in real-world education, where teachers work together to prepare lessons and students engage in peer learning. Building upon this concept, our innovative contributions in model construction are highlighted as follows: (1) A teacher knowledge augmentation module: this module is proposed to combine lower-level teacher features, facilitating the knowledge transfer from the teacher to the student. (2) A student mutual learning module is introduced to enable students to learn from each other, mimicking the peer learning concept in collaborative learning. (3) The Teacher Share Module combines lower-level teacher features: the specific functionality of the teacher knowledge augmentation module is described, which involves combining lower-level teacher features. (4) The multi-step transfer process can be easily optimized due to the minimal gap between the features: the proposed approach breaks down the knowledge transfer process into multiple steps, which can be easily optimized due to the minimal gap between the features involved in each step. Shared-KD uses simple feature losses without additional weights in transformation, resulting in an efficient distillation process that can be easily combined with other methods for further improvement. The effectiveness of our approach is validated through experiments on popular tasks such as object detection and instance segmentation.

Empowering Object Detection: Unleashing the Potential of Decoupled and Interactive Distillation

Research on Knowledge Distillation Algorithm of Object Detection

Hands-on Guidance for Distilling Object Detectors

Improve Object Detection with Feature-based Knowledge Distillation: Towards Accurate and Efficient Detectors.

Structured Knowledge Distillation for Accurate and Efficient Object Detection

Distilling Object Detectors with Global Knowledge

Instance-Aware Distillation for Efficient Object Detection in Remote Sensing Images

Distilling Object Detectors via Decoupled Features

Shared Knowledge Distillation Network for Object Detection

'Parallel-Circuitized' distillation for dense object detection

Teaching with Uncertainty: Unleashing the Potential of Knowledge Distillation in Object Detection

UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View

Improve Object Detection with Knowledge Distillation

Multilayer Semantic Features Adaptive Distillation for Object Detectors

PointDistiller: Structured Knowledge Distillation Towards Efficient and Compact 3D Detection

Learning Lightweight Object Detectors via Multi-Teacher Progressive Distillation

Distilling Object Detectors With Fine-Grained Feature Imitation

Adaptive Knowledge Distillation for Lightweight Remote Sensing Object Detectors Optimizing

Focal and Global Knowledge Distillation for Detectors

Exploring Effective Knowledge Distillation for Tiny Object Detection

Efficient Object Detection in Optical Remote Sensing Imagery via Attention-based Feature Distillation