Lemon Inc.

Beijing Zitiao Network Technology Co., Ltd.

Longyin Wen

Yufei Wang

Xinyao Wang

Ye Yuan

Fan Chen

Guang Chen

Sijie Zhu

Kai Xu

Xin Gu

Linjie Yang

Xiaojie Li

Ding Liu

Ziyu Jiang

Lu Xu

Zhiwei Deng

Congcong Li

Dexiang Hong

The present disclosure describes techniques for searching editing components based on text using a machine learning model. A plurality of visual embeddings indicative of a plurality of visual editing components is acquired by the machine learning model. The plurality of visual embeddings indicative of the plurality of visual editing components is projected into a common space by a first sub-model of the machine learning model. A text query input is received by a user. A text embedding indicative of the text query is generated. The text embedding is projected into the common space by a second sub-model of the machine learning model. At least one visual editing component among the plurality of visual editing components is determined based on the projected text embedding and the plurality of projected visual embeddings in the common space. Information indicative of the at least one visual editing component is displayed via a user interface.

Searching editing components based on text using a machine learning model

Methods, apparatuses and electronic devices for information segmentation. The method includes, for a target information node in an information sequence and based on first node information of the target information node, determine a first demarcation point probability value, in which the first demarcation point probability value indicates a probability that the target information node is a first type of demarcation point; determine, based on the first demarcation point probability value and second node information of the target information node, a second demarcation point probability value, in which the second demarcation point probability value indicates a probability that the information node is a second type of demarcation point, and the first type of demarcation point is associated with the second type of demarcation point; and determines, based on the first and the second demarcation point probability value, at least two segmentation modes for the information sequence.

Information segmentation methods, apparatuses, and electronic devices

The present disclosure describes techniques for generating representations of editing components using a machine learning model. Images and guidance tokens are input into a first sub-model of the machine learning model. The machine learning model is trained to distinguish the editing components from raw materials and generate the representations of the editing components. Tokens corresponding to the images are generated by the first sub-model based on the images and the guidance tokens. The tokens corresponding to the images and the guidance tokens are input into a second sub-model of the machine learning model. The second sub-model comprises a cross-attention mechanism. An embedding indicative of at least one editing component is generated based on the tokens corresponding to the images and the guidance tokens by the second sub-model.

Generating representations of editing components using a machine learning model

A method for adding animation to a video includes: detecting whether the hand of a user in the current video frame, which is currently captured, holds an object; if so, determining a target animation corresponding to the object; and determining an animation addition area, which corresponds to the target animation, on the current video frame, and adding the target animation to the animation addition area. Therefore, a corresponding animation special effect is added for an object held by a user in a video, such that the interaction between the user and the handheld object is visually reflected, thereby improving the interestingness of video filming, and providing technical support for improving the viscosity of the user with respect to a video product.

Method and apparatus for adding animation to video, and device and medium

Provided are a processing method and apparatus, an electronic device and a medium. The method includes steps described below. A target video is acquired; video information of the target video is determined; copy information corresponding to the video information is generated by using a video description model, where the copy information is generated by using the video description model and based on a task prompt and a copy keyword. Through this method, the video description model is constrained based on the task prompt and the copy keyword, so that the copy information of the target video is generated more accurately, and the coupling degree between the copy information and the target video is improved.

Processing method and apparatus, electronic device and medium

Provided are a processing method, a terminal device, and a medium. The method includes the steps described below. Basic information of a target video is determined; attribute information of the target video is determined based on the basic information; in a case where the attribute information indicates that the target video is a video capable of being structured, chapter division is performed on the target video based on the basic information to obtain at least two video clips; and chapter description information of the at least two video clips, a key frame of the at least two video clips and video description information of the target video are determined.

Processing method, terminal device, and medium

Provided are a video processing method and apparatus, an electronic device, and a medium. The method includes that a candidate slice information set of an original video is determined based on video information of the original video, the original video is segmented into a plurality of video segments based on each piece of candidate slice information in the candidate slice information set, the plurality of video segments is input into an evaluation model to determine a piece of probability information of each piece of candidate slice information, and target slice information of the original video is determined based on the piece of probability information corresponding to each piece of candidate slice information.

Video processing method, electronic device, and non-transitory computer-readable storage medium

A target video and video description information corresponding to the target video are acquired; salient object information of the target video is determined; a key frame category of the video description information is determined; and the target video, the video description information, the salient object information and the key frame category are input into a processing model to obtain a timestamp of an image corresponding to the video description information in the target video.

Processing method and apparatus, terminal device and medium

Systems and method directed to performing video object segmentation are provided. In examples, video data representing a sequence of image frames and video data representing an object mask may be received at a video object segmentation server. Image features may be generated based on a first image frame of the sequence of image frames, image features may be generated based on a second image frame of the sequence of image frames; and object features may be generated based on the object mask. A transform matrix may be computed based on the image features of the first image frame and image features of the second image frame; the transform matrix may be applied to the object features resulting in transformed object features. A predicted object mask associated with the second image frame may be obtained by decoding the transformed object features.

Growing community of inventors

Los Angeles, CA, United States of America

Longyin Wen