Object identification device and object identification method
AssigneeWistron Corporation
InventorKuo-Lun Huang
Life of this patent
Estimating the term…
Abstract
An object identification method includes: generating a tracking sample and an adversarial sample; training a teacher model according to the tracking sample; and initializing a student model according to the teacher model. The student model adjusts a plurality of parameters according to the teacher model and the adversarial sample, in response to the vector difference between the output result of the student model and the output result of the teacher model being lower than the learning threshold, the student model is deemed to have completed training, and the student model is extracted as an object identification model.
Claims20
An object identification device, comprising:
- a processor;
- and
- a storage device;
- wherein the processor is configured to access programs stored in the storage device to implement a pre-processing module, a teacher model training module, and a student model training module;
- wherein
- the pre-processing module is configured to generate a tracking sample and an adversarial sample;
- the teacher model training module is configured to use the tracking sample to train a teacher model;
- and
- the student model training module is configured to initialize a student model according to the teacher model;
- wherein the student model adjusts a plurality of parameters according to the teacher model and the adversarial sample, in response to a vector difference between an output result of the student model and an output result of the teacher model being less than a learning threshold, the student model is deemed to have completed training, and the student model is extracted as an object identification model.
An object identification method, comprising:
- generating a tracking sample and an adversarial sample;
- training a teacher model according to the tracking sample;
- and
- initializing a student model according to the teacher model;
- wherein the student model adjusts a plurality of parameters according to the teacher model and the adversarial sample, and in response to a vector difference between an output result of the student model and an output result of the teacher model being lower than a learning threshold, the student model is deemed to have completed training, and the student model is extracted as an object identification model.
Dependencies are inferred from the claim text.