Method for training object detection model, object detection method and apparatus

A method for training an object detection model, the method comprising:

• obtaining M sample image sets, wherein a sample image set includes at least one sample image and at least one object type of at least one object in each sample image, an object type corresponds to one sample image set, the M sample image sets correspond to N object types, M is a positive integer less than or equal to N, and N is a positive integer greater than 1;
• obtaining an initial object detection model, wherein the initial object detection model includes M network layers or an object detection network;
• the M network layers each are used for detecting at least one object of at least one object type;
• a number of output channels of the object detection network is N, and the object detection network is used for detecting objects of N object types;
• and
• training the initial object detection model by using the M sample image sets to obtain the object detection model.