Embodiments of the present application provide a model training method, device, apparatus, and storage medium. The method includes: acquiring a target image sample data set comprising a first image sample including a target object and a second image sample excluding the target object; obtaining a target model for identifying the target object and identifying whether the target object is included in an image based on train of the first image sample and the second image sample. Embodiments of that present application can improve the accuracy of a model detection object.