Object detection model training method and apparatus, object detection method and apparatus, computer device, and storage medium

An object detection model training apparatus, comprising:

• at least one memory configured to store program code;
• at least one processor configured to read the program code and operate as instructed by the program code, the program code comprising:
• prediction code configured to cause the at least one processor to input a first sample image into an initial detection model of a current round, and output a first prediction result for a target object, the first sample image not carrying position annotation information of the target object;
• transformation code configured to cause the at least one processor to transform the first sample image and a first prediction position region of the target object in the first prediction result, to obtain a second sample image and a prediction transformation result for the target object in the second sample image,
• the prediction code being further configured to cause the at least one processor to input the second sample image into the initial detection model, and output a second prediction result for the target object;
• unsupervised loss determining code configured to cause the at least one processor to obtain a loss value of unsupervised learning according to a difference between the second prediction result and the prediction transformation result;
• and
• parameter adjustment code configured to cause the at least one processor to adjust model parameters of the initial detection model according to the loss value, notify the prediction module to use a next round as the current round, and continue to perform the operation of inputting a first sample image into an initial detection model of a current round to perform iterative training until a training end condition is met, to obtain an object detection model.

A non-transitory computer-readable storage medium, storing computer program code that when executed by at least one processor causes the at least one processor to:

• input a first sample image into an initial detection model of a current round, and outputting a first prediction result for a target object, the first sample image not carrying position annotation information of the target object;
• transform the first sample image and a first prediction position region of the target object in the first prediction result to obtain a second sample image and a prediction transformation result for the target object in the second sample image;
• input the second sample image into the initial detection model, and output a second prediction result for the target object;
• obtain a loss value of unsupervised learning according to a difference between the second prediction result and the prediction transformation result;
• and
• adjust model parameters of the initial detection model according to the loss value, using a next round as the current round, and return to the operation of inputting a first sample image into an initial detection model of a current round to perform iterative training until a training end condition is met, to obtain an object detection model.