Electronic device and method for preventing non-fungible token plagiarism in electronic device
According to an embodiment, an electronic device comprises: a communication circuit, at least one processor, and memory storing instructions. The electronic device may convert a first image into a latent vector using an encoder. The electronic device may obtain a first hash value by applying a hashing method to the latent vector for the first image. The electronic device may compare the first hash value for the first image with a second hash value for a second image; and determine whether the first image is plagiarized based on a result of the comparison.
1 . An electronic device, comprising:
a communication circuit;
at least one processor, comprising processing circuitry; and
memory storing instructions for execution by at least one processor, wherein at least one processor, individually and/or collectively, is configured to cause the electronic device to:
convert a first image into a latent vector using an encoder;
obtain a first hash value by applying a hashing method to the latent vector for the first image; and
compare the first hash value for the first image with a second hash value for a second image and determine whether the first image is plagiarized based on a result of the comparison.
2 . The electronic device of claim 1 , wherein the electronic device is configured to:
based on determining that the first image does not plagiarize other images, generate a non-fungible token (NFT) for the first image including the first hash value.
3 . The electronic device of claim 1 , wherein the hashing method includes locality sensitive hashing (LSH).
4 . The electronic device of claim 1 , wherein the electronic device is configured to:
perform quantization on the latent vector of the first image; and
apply the hashing method to the quantized latent vector.
5 . The electronic device of claim 1 , wherein the electronic device is configured to set as a section for the quantization a minimum section where a change in an image output from a decoder occurs due to a change in a feature for the first image in the encoder.
6 . The electronic device of claim 1 , wherein a number of dimensions of the latent vector generated using the encoder is 128 or more.
7 . The electronic device of claim 1 , wherein the electronic device is configured to:
determine whether a similarity between the first hash value for the first image and the second hash value for the second image is a threshold or more; and
based on the similarity being the threshold or more, identify whether the first image plagiarizes the second image.
8 . The electronic device of claim 1 , wherein the electronic device is configured to generate a list including at least one image having a similarity to the first image, having the threshold or more.
9 . A method of operating an electronic device, comprising:
converting a first image into a latent vector using an encoder;
obtaining a first hash value by applying a hashing method to the latent vector for the first image; and
comparing the first hash value for the first image with a second hash value for a second image and determining whether the first image is plagiarized based on a result of the comparison.
10 . The method of claim 9 , further comprising, based on determining that the first image does not plagiarize other images, generating a non-fungible token (NFT) for the first image including the first hash value.
11 . The method of claim 9 , wherein the hashing method includes locality sensitive hashing (LSH).
12 . The method of claim 9 , further comprising:
performing quantization on the latent vector of the first image; and
applying the hashing method to the quantized latent vector.
13 . The method of claim 9 , wherein a section for the quantization is set as a minimum section where a change in an image output from a decoder occurs due to a change in a feature for the first image in the encoder.
14 . The method of claim 9 , wherein a number of dimensions of the latent vector generated using the encoder is 128 or more.
15 . The method of claim 9 , further comprising:
determining whether a similarity between the first hash value for the first image and the second hash value for the second image is a threshold or more; and
based on the similarity being the threshold or more, identifying whether the first image plagiarizes the second image.
16 . The method of claim 9 , further comprising
generating a list including at least one image having a similarity to the first image, having the threshold or more.
17 . A non-transitory computer-readable storage medium storing at least one instruction,
wherein the at least one instruction, when executed by at least one processor, comprising processing circuitry, individually and/or collectively, cause an electronic device to perform operations, comprising:
converting a first image into a latent vector using an encoder;
obtaining a first hash value by applying a hashing method to the latent vector for the first image; and
comparing the first hash value for the first image with a second hash value for a second image and determining whether the first image is plagiarized based on a result of the comparison.
18 . The non-transitory computer-readable storage medium of claim 17 , wherein the plurality of operations further comprise, based on determining that the first image does not plagiarize other images, generating a non-fungible token (NFT) for the first image including the first hash value.
19 . The non-transitory computer-readable storage medium of claim 17 , wherein the hashing method includes locality sensitive hashing (LSH).
20 . The non-transitory computer-readable storage medium of claim 17 , wherein the plurality of operations further comprise:
performing quantization on the latent vector of the first image; and
applying the hashing method to the quantized latent vector.