Method and apparatus for detecting target object

1. A method of detecting a target object performed by a computing device including at least one processor, the method comprising:

receiving an input image; and

generating first result information related to a target object of a plurality of target objects from the input image based on a neural network-based detection model trained based on a training data set including at least one training image and Ground Truth (GT) data corresponding to each training image,

wherein the GT data includes a channel including adjacent area information about an area adjacent between the target object and another of the plurality of target objects, wherein the channel comprises a data set having a value allocated to at least one pixel included in the input image.

2. The method of claim 1 , wherein the training data set is generated based on an incremental operation that generates a plurality of new training image-GT data pairs from at least one existing training-image-GT data pair, and is generated before being input to the detection model for training the detection model.

3. The method of claim 1 , wherein the GT data further includes a channel including contour information about the target object.

4. The method of claim 1 , wherein the GT data further includes:

a channel including contour information about the target object; and

a channel including target area information.

5. The method of claim 1 , wherein the generating of the first result information includes generating single-channel output data based on multi-channel output data generated from the input image by the detection model.

6. The method of claim 5 , wherein the generating of the single-channel output data includes adjusting a value included in a third channel based on at least one of a first channel and a second channel of the multi-channel output data, and

the multi-channel output data includes:

a first channel including contour information about the target object;

a second channel including adjacent area information about an area adjacent between two different target objects; and

a third channel including target area information.

7. The method of claim 6 , wherein the adjusting includes:

changing at least a part of values allocated to an area corresponding to a contour determined based on the first channel to a background value in the third channel; and

changing at least a part of values allocated to an area corresponding to an adjacent area determined based on the second channel to a background value in the third channel.

8. The method of claim 1 , further comprising:

generating a first polarized image by performing a first decomposition operation on the input image; and

generating a synthesized image through an image generating model based on the input image,

wherein the generating of the first result information includes generating the first result information related to the target object from the input image additionally based on the first polarized image and the synthesized image.

9. The method of claim 1 , further comprising:

generating second result information related to an individual area of each of the two or more target objects that do not overlap each other within the input image based on the first result information.

10. The method of claim 9 , further comprising:

calculating individual contour information about each of the two or more target objects based on the second result information.

11. A non-transitory computer readable storage medium storing a computer program, wherein when the computer program is executed by one or more processors, the computer program causes following operations for detecting a target object to be performed, the operations comprising:

receiving an input image; and

generating first result information related to a target object of a plurality of target objects from the input image based on a neural network-based detection model trained based on a training data set including at least one training image and Ground Truth (GT) data corresponding to each training image, and

the GT data includes a channel including adjacent area information about an area adjacent between the target object and another of the plurality of target objects, wherein the channel comprises a data set having a value allocated to at least one pixel included in the input image.

12. An apparatus for detecting a target object, the apparatus comprising:

one or more processors;

one or more memories; and

a network unit,

wherein the one or more processors receive an input image, and generate first result information related to a target object of a plurality of target objects from the input image based on a neural network-based detection model trained based on a training data set including at least one training image and Ground Truth (GT) data corresponding to each training image, and

the GT data includes a channel including adjacent area information about an area adjacent between the target object and another of the plurality of target objects, wherein the channel comprises a data set having a value allocated to at least one pixel included in the input image.