REVIEW 2 cited by
CrossKD: Cross-Head Knowledge Distillation for Object Detection
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Knowledge Distillation (KD) has been validated as an effective model compression technique for learning compact object detectors. Existing state-of-the-art KD methods for object detection are mostly based on feature imitation. In this paper, we present a general and effective prediction mimicking distillation scheme, called CrossKD, which delivers the intermediate features of the student's detection head to the teacher's detection head. The resulting cross-head predictions are then forced to mimic the teacher's predictions. This manner relieves the student's head from receiving contradictory supervision signals from the annotations and the teacher's predictions, greatly improving the student's detection performance. Moreover, as mimicking the teacher's predictions is the target of KD, CrossKD offers more task-oriented information in contrast with feature imitation. On MS COCO, with only prediction mimicking losses applied, our CrossKD boosts the average precision of GFL ResNet-50 with 1x training schedule from 40.2 to 43.7, outperforming all existing KD methods. In addition, our method also works well when distilling detectors with heterogeneous backbones. Code is available at https://github.com/jbwang1997/CrossKD.
Forward citations
Cited by 2 Pith papers
-
HKT: A Biologically Inspired Framework for Modular Hereditary Knowledge Transfer in Neural Networks
HKT is a modular feature-level distillation method whose genetic attention residual improves compact vision models on optical flow, classification, and segmentation benchmarks.
-
Knowledge Distillation for Reservoir-based Classifier: Human Activity Recognition
PatchEchoClassifier combines an echo state network with DeiT-style distillation from an MLP-Mixer teacher to reach about 86% accuracy on SHL human activity data at much lower FLOPs, though it falls below 80% on four o...
Discussion (0). Continue with ORCID to comment.