Method and electronic device for object recognition, and method for acquiring depth information of an object

A method for acquiring depth information of an object from two original images, which are captured at the same time respectively by two cameras that are spaced apart from each other and that have an overlapping field of view containing the object, the method to be implemented by an electronic device and comprising the steps of:

• (a) configuring the electronic device to access the two original images;
• (b) configuring the electronic device to reduce resolutions of the two original images so as to generate two resolution-reduced images, respectively;
• (c) configuring the electronic device to calculate a plurality of shift amounts, each of which is between two corresponding pixels in pixel blocks that have similar content and that are respectively in the two resolution-reduced images, and to generate a low-level depth image based on the shift amounts, a larger one of the shift amounts representing a shallower depth on the low-level depth image;
• (d) configuring the electronic device to determine an object area of the low-level depth image containing the object therein;
• (e) configuring the electronic device to obtain two sub-images respectively, from the two original images, corresponding to the object area of the low-level depth image determined in step (d);
• and
• (f) configuring the electronic device to calculate a plurality of shift amounts, each of which is between two corresponding pixels in pixel blocks that have similar content and that are respectively in the two sub-images obtained in step (e), and to generate a high-level depth image of the object to serve as the depth information based on the shift amounts, a larger one of the shift amounts representing a shallower depth on the high-level depth image.

An electronic device for recognizing an object from two original images, which are captured at the same time respectively by two cameras that are spaced apart from each other and that have an overlapping field of view containing the object, said electronic device comprising:

• an input unit adapted to receive the original images;
• a storage unit storing program instructions which are associated with a method for recognizing an object;
• and
• a processor coupled to said input unit and said storage unit, and configured to execute the program instructions stored in said storage unit so as to perform the following steps of
• (i) accessing the two original images,
• (ii) reducing resolutions of the two original images so as to generate two resolution-reduced images, respectively,
• (iii) calculating a plurality of shift amounts, each of which is between two corresponding pixels in pixel blocks that have similar content and that are respectively in the two resolution-reduced images, and generating a low-level depth image based on the shift amounts, a larger one of the shift amounts representing a shallower depth on the low-level depth image,
• (iv) determining an object area of the low-level depth image containing the object therein, and
• (v) obtaining a sub-image, from one of the original images, corresponding to the object area of the low-level depth image, and storing the sub-image in said storage unit, thereby recognizing the object based on the sub-image.

A method for acquiring a sub-image containing an object from two original images, which are captured at the same time respectively by two cameras that are spaced apart from each other and that have an overlapping field of view containing the object, the method to be implemented by an electronic device and comprising the following steps of:

• (A) configuring the electronic device to obtain two to-be-compared images respectively from the two original images, wherein at least one of the to-be-compared images is generated by reducing resolution of a corresponding one of the original images;
• (B) configuring the electronic device to calculate a plurality of shift amounts, each of which is between two corresponding pixels in pixel blocks that have similar content and that are respectively in the two to-be-compared images, and to generate a low-level depth image based on the shift amounts, a larger one of the shift amounts representing a shallower depth on the low-level depth image;
• (C) configuring the electronic device to determine an object area of the low-level depth image containing the object therein;
• and
• (D) configuring the electronic device to obtain the sub-image, from at least one of the original images, corresponding to the object area of the low-level depth image determined in step (C).

An electronic device for acquiring a sub-image containing an object from two original images, which are captured at the same time respectively by two cameras that are spaced apart from each other and that have an overlapping field of view containing the object, said electronic device comprising:

• an input unit adapted to receive the original images;
• and
• a processor coupled to said input unit and configured to
• (i) obtain two to-be-compared images respectively from the two original images, wherein at least one of the to-be-compared images is generated by reducing resolution of a corresponding one of the original images;
• (ii) calculate a plurality of shift amounts, each of which is between two corresponding pixels in pixel blocks that have similar content and that are respectively in the two to-be-compared images, and generate a low-level depth image based on the shift amounts, a larger one of the shift amounts representing a shallower depth on the low-level depth image,
• (iii) determine an object area of the low-level depth image containing the object therein, and
• (iv) obtain the sub-image, from at least one of the original images, corresponding to the object area of the low-level depth image determined in step (iii).