Abstract: A mobile application with an improved user interface facilitates generating stylized media content items including images and videos. An end-user selects a desired visual effect from a set of options. The mobile application captures or accesses an image. The image is processed on a server using a generative neural network pre-trained to apply stylizations based on the selected effect. The server sends back the stylized image to the mobile application for display. The end-user can then save the stylized image or generate a video (e.g., an animation) showing the original image transition to the stylized image. The user interface provides an efficient creative workflow to apply aesthetic enhancements in a visual style chosen by the end-user. Generative machine learning techniques automate stylization to enable accessible media customization and sharing.

Abstract: Methods and systems are disclosed for enhancing or modifying an image by a diffusion model. The methods and systems receive a first image depicting a real-world scene including a target object and receive input associated with adjusting a zoom level of the first image. The methods and systems, in response to receiving the input, modify the zoom level associated with the first image to generate a second image having a view of the target object that is different from a view of the target object in the first image. The methods and systems analyze the second image using a generative machine learning model to generate an artificial image that modifies portions of the second image to improve the view of the target object relative to the second image.

Abstract: Examples described herein relate to techniques for facilitating selection of stickers for inclusion in messages within the context of an interaction system. According to some examples, message content is detected and a set of candidate stickers is identified based on the message content. A search icon is dynamically replaced with a representation of respective ones of the set of candidate stickers. At a first point in time, the search icon represents a first candidate sticker of the set of candidate stickers. At a second point in time, the search icon represents a second candidate sticker of the set of candidate stickers.

Abstract: A mobile application with an improved user interface facilitates generating stylized media content items including images and videos. An end-user selects a desired visual effect from a set of options. The mobile application captures or accesses an image. The image is processed on a server using a generative neural network pre-trained to apply stylizations based on the selected effect. The server sends back the stylized image to the mobile application for display. The end-user can then save the stylized image or generate a video (e.g., an animation) showing the original image transition to the stylized image. The user interface provides an efficient creative workflow to apply aesthetic enhancements in a visual style chosen by the end-user. Generative machine learning techniques automate stylization to enable accessible media customization and sharing.

Abstract: Methods and systems are disclosed for enhancing or modifying an image by a diffusion model. The methods and systems receive a first image depicting a real-world scene including a target object and receive input associated with adjusting a zoom level of the first image. The methods and systems, in response to receiving the input, modify the zoom level associated with the first image to generate a second image having a view of the target object that is different from a view of the target object in the first image. The methods and systems analyze the second image using a generative machine learning model to generate an artificial image that modifies portions of the second image to improve the view of the target object relative to the second image.

Abstract: Examples described herein relate to techniques for facilitating selection of stickers for inclusion in messages within the context of an interaction system. According to some examples, sticker selection history data of a user of an interaction application is retrieved. Absence of a sticker suggestion trigger with respect to a messaging interface of the interaction application is detected. Presence of the sticker suggestion trigger results in presentation of a suggested sticker for transmission via the messaging interface. Based on the sticker selection history data and the absence of the sticker suggestion trigger, a sticker search icon state is selected. A search icon corresponding to the sticker search icon state is caused to be presented within the messaging interface.

Abstract: Examples described herein relate to techniques for facilitating selection of stickers for inclusion in messages within the context of an interaction system. According to some examples, message content is detected and a set of candidate stickers is identified based on the message content. A search icon is dynamically replaced with a representation of respective ones of the set of candidate stickers. At a first point in time, the search icon represents a first candidate sticker of the set of candidate stickers. At a second point in time, the search icon represents a second candidate sticker of the set of candidate stickers.

Abstract: A method includes determining participation in an interaction function by a first user of an interaction system with a second user of the interaction system. The method also includes accessing profile data of the first user, and determining, based on the profile data, whether the first user has captured or designated a first-user self-image for use in the interaction function. In response to determining that the first user has not captured or designated the first-user self-image, the method includes accessing a media content item that includes a character, identifying a head portion of the character in the media content item, replacing the head portion with a placeholder space, and displaying the media content item with the placeholder space in a user interface corresponding to the interaction function.

Abstract: A method includes determining participation in an interaction function by a first user of an interaction system with a second user of the interaction system. The method also includes accessing profile data of the first user, and determining, based on the profile data, whether the first user has captured or designated a first-user self-image for use in the interaction function. In response to determining that the first user has not captured or designated the first-user self-image, the method includes accessing a media content item that includes a character, identifying a head portion of the character in the media content item, replacing the head portion with a placeholder space, and displaying the media content item with the placeholder space in a user interface corresponding to the interaction function.

Abstract: Examples in the present disclosure relate to prompt-driven, conversation-specific image generation. A first user device of a first user of an interaction application transmits an image generation request. The interaction application enables the first user to interact with at least a second user. An image is automatically generated based on a prompt included in the image generation request, and the image is presented at the first user device. In response to receiving user input to select the image, the image is stored as a conversation-specific image in association with a first user profile of the first user and a second user profile of the second user. The conversation-specific image is presented at both the first user device of the first user and a second user device of the second user, together with conversation data associated with a conversation between the first user and the second user.

Abstract: Provided are systems and methods for searching and ranking personalized videos. An example method includes receiving a user selection of a category from a set of pre-determined categories, selecting, from a pool of videos, a first subset of videos, where a first video from the first subset of videos is associated with the category and having a calculated value of a performance metric, selecting, from the pool of videos, a second subset of videos, where a second video from the second subset of videos is associated with the category and having an unknown value for the performance metric, generating, based on a predetermined rule, a third set of videos, where the third set of videos includes at least one video from the first subset of videos and at least one video from the second subset of videos, and presenting the third set of videos via a communication chat.

Abstract: Provided are systems and methods for customizing modifiable videos. An example method includes analyzing recent messages associated with a user in a multimedia messaging application to determine an emotional state of the user, determining, based on the emotional state, a customized property of a modifiable feature of a modifiable video, where the modifiable video includes a preset property of the modifiable feature, replacing the preset property of the modifiable feature with the customized property of the modifiable feature, providing a user interface enabling the user to view the modifiable video and modify the customized property of the modifiable feature to a modified property, and, upon determining that the user has modified the customized property, storing information concerning the emotional state, the customized property, and the modified property to a statistical log.

Abstract: Systems and methods for searching and ranking personalized videos are provided. An example method includes receiving a user request via a communication chat between a user of a computing device and a further user of a further computing device, searching, based on the user request, a pool of personalized videos to determine a first subset of personalized videos, determining first rankings of personalized videos in the first subset of personalized videos, the first rankings being based on global statistical data associated with the personalized videos, selecting, based on the first rankings, a second subset of personalized videos from the first subset of personalized videos, determining, based on at least one characteristic associated with the user, second rankings of personalized videos from the second subset of personalized videos, and presenting the personalized videos selected from the second subset within the communication chat and in order according to the second rankings.

Abstract: Various examples described herein support or provide generation and management operations of personalized images using machine learning technologies, including receiving portrait images of an entity; using machine learning models to generate an identity that represents the entity; identifying a template that comprises text descriptions of a scene and conditions; and using a machine learning models to generate a personalized image based on the identity and the template.

Abstract: Provided are systems and methods for customizing modifiable videos. An example method includes analyzing recent messages associated with a user in a multimedia messaging application to determine a context of the recent messages, determining, based on the context, a property of a modifiable feature, selecting, based on the context, a list of relevant modifiable videos from a database configured to store modifiable videos associated with a preset modifiable feature, replacing a property of the preset modifiable feature in relevant modifiable videos of the list of relevant modifiable videos with the property of the modifiable feature, and rendering the list of relevant modifiable videos for viewing by the user, where the rendering includes displaying the modifiable feature in the relevant modifiable videos.

Abstract: Examples disclosed herein describe techniques related to automated image generation in an interaction system. An image generation request is received from a first user device associated with a first user of an interaction system. The image generation request comprises a text prompt. Responsive to receiving the image generation request, an image is automatically generated by an automated text-to-image generator, based on the text prompt. The image is caused to be presented on the first user device. An indication of user input to select the image is received from the user device. Responsive to receiving the indication of the user input to select the image, the image is associated with the first user within the interaction system, and a second user of the interaction system is enabled to be presented with the image.

Abstract: A method includes determining participation in an interaction function by a first user of an interaction system with a second user of the interaction system. The method also includes accessing profile data of the first user, and determining, based on the profile data, whether the first user has captured or designated a first-user self-image for use in the interaction function. In response to determining that the first user has not captured or designated the first-user self-image, the method includes accessing a media content item that includes a character, identifying a head portion of the character in the media content item, replacing the head portion with a placeholder space, and displaying the media content item with the placeholder space in a user interface corresponding to the interaction function.

Abstract: Methods and systems are disclosed for enhancing or modifying an image by a diffusion model. The methods and systems receive a first image depicting a real-world scene including a target object and receive input associated with adjusting a zoom level of the first image. The methods and systems, in response to receiving the input, modify the zoom level associated with the first image to generate a second image having a view of the target object that is different from a view of the target object in the first image. The methods and systems analyze the second image using a generative machine learning model to generate an artificial image that modifies portions of the second image to improve the view of the target object relative to the second image.

Abstract: Examples described herein relate to techniques for facilitating selection of stickers for inclusion in messages within the context of an interaction system. According to some examples, sticker selection history data of a user of an interaction application is retrieved. Absence of a sticker suggestion trigger with respect to a messaging interface of the interaction application is detected. Presence of the sticker suggestion trigger results in presentation of a suggested sticker for transmission via the messaging interface. Based on the sticker selection history data and the absence of the sticker suggestion trigger, a sticker search icon state is selected. A search icon corresponding to the sticker search icon state is caused to be presented within the messaging interface.

Abstract: Examples described herein relate to techniques for facilitating selection of stickers for inclusion in messages within the context of an interaction system. According to some examples, message content is detected and a set of candidate stickers is identified based on the message content. A search icon is dynamically replaced with a representation of respective ones of the set of candidate stickers. At a first point in time, the search icon represents a first candidate sticker of the set of candidate stickers. At a second point in time, the search icon represents a second candidate sticker of the set of candidate stickers.