Abstract: In particular embodiments, a method includes, by one or more computing devices, receiving an image requiring an administrative action. The method includes, by the one or more computing devices, generating a single large hash value based on the image and a content-aware hashing function. The large hash value has a first number of dimensions. The method includes, by the one or more computing devices, generating a small hash value corresponding to the image from the large hash value using a locality-sensitive hashing function. The small hash value has a second number of dimensions that is smaller than the first number of dimensions. The method includes determining, based on the small hash value corresponding to the image, a cluster center mutually associated with the image. The method includes, by the one or more computing devices, performing the administrative action on one or more images mutually associated with the cluster center.

Abstract: In particular embodiments, a method includes, by one or more computing devices, receiving an image requiring an administrative action. The method includes, by the one or more computing devices, generating a single large hash value based on the image and a content-aware hashing function. The large hash value has a first number of dimensions. The method includes, by the one or more computing devices, generating a small hash value corresponding to the image from the large hash value using a locality-sensitive hashing function. The small hash value has a second number of dimensions that is smaller than the first number of dimensions. The method includes determining, based on the small hash value corresponding to the image, a cluster center mutually associated with the image. The method includes, by the one or more computing devices, performing the administrative action on one or more images mutually associated with the cluster center.

Abstract: In one embodiment, a computing device receives an image. A large hash value is generated for the image based on a content-aware hashing function, resulting in the large hash value having a first number of dimensions. A small hash value is calculated for the image based on the large hash value and a locality-sensitive hashing function. The small hash value has a reduced number of dimensions compared to the large hash value. The small hash values for two or more images, respectively, will be close to each other if the two or more images are similar in content, and the small hash values will be different if the images are not similar. The small hash value for the image is used to determine whether the image is similar to one or more other images.

Abstract: In one embodiment, a computing device receives an image. A large hash value is generated for the image based on a content-aware hashing function, resulting in the large hash value having a first number of dimensions. A small hash value is calculated for the image based on the large hash value and a locality-sensitive hashing function. The small hash value has a reduced number of dimensions compared to the large hash value. The small hash values for two or more images, respectively, will be close to each other if the two or more images are similar in content, and the small hash values will be different if the images are not similar. The small hash value for the image is used to determine whether the image is similar to one or more other images.

Abstract: In one embodiment, a computing device receives a first image from a client device associated with at least one user of a social-networking system. The computing device performs a content-aware hashing function on the first image and generates a large hash value. The computing device then performs a locality-sensitive hashing function on the large hash value to generate a small hash value. The computing device calculates a distance from the small hash value to a cluster center which is associated with the small hash values for at least one other image. If the distance is greater than a threshold distance, the first image is determined not similar to the at least one other image. The computing device creates a new cluster center for the first image. If the distance is less than the threshold distance, the first image is determined similar to the at least one other image.

Abstract: In one embodiment, a method includes determining by a computer system a distance between a first value associated with a first image and a first cluster center associated with at least one image and determining by the computer system the first image and the at least one image are not similar when the determined distance is more than a threshold distance.

Abstract: In one embodiment, a method includes determining by a computer system a distance between a first value associated with a first image and a first cluster center associated with at least one image and determining by the computer system the first image and the at least one image are not similar when the determined distance is more than a threshold distance.