Abstract: Disclosed is a method for performing object segmentation using a neural network model, the method performed by one or more processors of a computing device according to an exemplary embodiment of the present disclosure. the method may include: obtaining a training image containing an object; training a neural network model to perform object segmentation using the training image; identifying a boundary region of the object included in the training image; obtaining an additional training image based on the training image and the identified boundary region; and additionally training the trained neural network model based on the additional training image.

Abstract: Disclosed is a method for anomaly detection using depth information, the method performed by one or more processors of a computing device according to an exemplary embodiment of the present disclosure. The method may include: obtaining first image data including depth information by scanning a surface of a product; generating reference image data of the product based on the first image data; obtaining second image data based on a difference between the reference image data and the first image data; and detecting a defect region of the product based on the second image data.

Abstract: Disclosed is an apparatus for inspecting an appearance of a product. The apparatus may include: a reference camera for capturing an object in a reference direction; and a plurality of auxiliary cameras for capturing the object in a plurality of auxiliary directions different from the reference direction and the plurality of auxiliary cameras may be arranged in a shape in which the plurality of auxiliary cameras are symmetric to each other when projected onto a plane perpendicular to the reference direction.

Abstract: Disclosed is a method for obtaining training data using a neural network model, the method is performed by one or more processors of a computing device according to an exemplary embodiment of the present disclosure. The method may include: obtaining a training image including an object having a first defect; obtaining a predicted image including a second defect using a neural network model; calculating a plurality of loss functions based on the training image and the predicted image; and calculating a final loss function based on the plurality of loss functions; and training the neural network model based on the final loss function.