This topic describes the product and documentation updates for Alibaba Cloud Visual Intelligence API in 2020.
December 2020
Category | Interface Name | Description | Release date | Region | References |
Image recognition | ID photo quality assessment | Detects whether an ID photo has quality issues and identifies them. | 2020-12-31 | China (Shanghai) | |
Video production | Video SDR Color Grading | Automatically optimizes the color of Standard Dynamic Range (SDR) videos based on content semantics and color to improve video color quality. | 2020-12-31 | China (Shanghai) | Unpublished |
SDR to HDR conversion | Converts standard SDR videos to High Dynamic Range (HDR) videos. This feature expands the color gamut to BT.2020, increases the color depth to 10 bit, and raises the brightness to a maximum of 1,000 nits to deliver higher quality video content. | 2020-12-31 | China (Shanghai) | Unpublished | |
Video frame interpolation | Uses a deep learning-based frame rate up-conversion method. It synthesizes video frames at any point in time through a frame interpolation network to fix video quality issues such as stuttering and jitter. | 2020-12-31 | China (Shanghai) | ||
Image analysis and processing | Rib fracture detection | Provides auxiliary diagnosis of rib fractures based on chest CT scans. It outputs the location and type of the fracture. | 2020-12-31 | China (Shanghai) | |
Chest CT scan screening | Performs detection and quantitative analysis of multiple organs and diseases in the human chest based on conventional chest CT images. | 2020-12-31 | China (Shanghai) |
November 2020
Category | Interface Name | Description | Release date | Region | References |
Image production | Image colorization | Automatically colorizes black and white photos and images. | 2020-11-30 | China (Shanghai) | |
Video production | Comprehensive video enhancement | Uses an AI deep learning algorithm to perform comprehensive enhancement processing on input SDR videos, including frame interpolation, super-resolution (SR), and SDR to HDR conversion. | 2020-11-30 | China (Shanghai) | |
Video face fusion | Fuses the facial features of one person into a specified face in a video to create a face-swapping effect. | 2020-11-30 | China (Shanghai) |
October 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Human anime stylization | Transforms a human image into a two-dimensional anime-style character and returns the stylized image. | 2020-10-30 | China (Shanghai) | |
Crowd heatmap estimation | Estimates the number of people in an image using a heatmap. | 2020-10-30 | China (Shanghai) | Unpublished | |
Image production | Human body erasing | Erases human figures from specified areas in an image and automatically fills the background. | 2020-10-30 | China (Shanghai) | |
OCR | Recaptured certificate detection | Detects whether a photo of a People's Republic of China resident ID card is a recaptured image from a screen. | 2020-10-30 | China (Shanghai) | |
Storefront recognition | Recognizes images of storefront signs and extracts information, such as the storefront photo, logo, store address, and contact information. | 2020-10-30 | China (Shanghai) | Unpublished |
September 2020
Category | API Name | Description | Release date | Region | References |
Face and body | Structured human attributes | Detects human attributes in an image. This feature includes human body detection and attribute prediction. | 2020-09-30 | China (Shanghai) | |
Image production | Image jitter | Animates static areas in an input image, such as the sky and a person's hair, to create a cinemagraph video in AVI format. | 2020-09-30 | China (Shanghai) | Unpublished |
Segmentation and matting | Skin segmentation | Recognizes and segments the skin areas of people in an image. | 2020-09-30 | China (Shanghai) | |
Image analysis and processing | Femoral neck fracture classification | Detects whether the femoral necks on both sides are fractured in an input anteroposterior hip X-ray. | 2020-09-30 | China (Shanghai) | Unpublished |
Preoperative knee measurement | Detects the coordinates and position labels of key points in an input anteroposterior full-length lower limb X-ray. | 2020-09-30 | China (Shanghai) | Unpublished | |
Coronary artery calcium scoring | Calculates the coronary artery calcium score based on a non-contrast chest CT scan. | 2020-09-30 | China (Shanghai) | ||
Preoperative hip measurement | Detects the coordinates and position labels of key points in an input anteroposterior hip X-ray. | 2020-09-30 | China (Shanghai) | Unpublished | |
Chest CT registration | Performs image registration on chest CT scans of the same patient taken at different times. | 2020-09-30 | China (Shanghai) | ||
Medical AI chat | Provides a medical and health Q&A service for pediatric disease education. It offers answers to common questions and similar questions. | 2020-09-30 | China (Shanghai) | ||
Skin disease detection | Performs skin disease classification and prediction on input natural images of pediatric skin. | 2020-09-30 | China (Shanghai) |
August 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Face data masking | Blurs faces in an input image and outputs the masked image. | 2020-08-31 | China (Shanghai) | |
Video segmentation | Half-body portrait segmentation | Segments the upper body portrait of a person in a video. | 2020-08-31 | China (Shanghai) | Unpublished |
Image analysis and processing | Knee X-ray KL grading | Analyzes an input knee X-ray image to determine the severity of arthritis. It returns the Kellgren-Lawrence (KL) grade and the knee joint position. | 2020-08-31 | China (Shanghai) | Unpublished |
Lumbar MRI qualitative analysis | Performs intelligent analysis on input DICOM images of the spine or lumbar region. It outputs information about intervertebral discs and vertebral bodies. | 2020-08-31 | China (Shanghai) | Unpublished | |
Medical machine translation | Performs machine translation on the input text for medical scenarios. | 2020-08-31 | China (Shanghai) | Unpublished |
July 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Celebrity recognition | Recognizes celebrities in an image. | 2020-07-31 | China (Shanghai) | |
Segmentation and matting | Logo segmentation | Separates the logo from an image and returns the segmented logo as a transparent PNG image. | 2020-07-31 | China (Shanghai) | Unpublished |
Outdoor scene segmentation | Performs pixel-level matting on scenes in an image. | 2020-07-31 | China (Shanghai) | Unpublished | |
Video production | Video aspect ratio transformation | Intelligently clips and fills an input video to output a video of any resolution. | 2020-07-31 | China (Shanghai) |
June 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Action and behavior recognition | Recognizes human actions and behaviors in videos and images and returns the identified behavior categories. | 2020-06-30 | China (Shanghai) | |
Segmentation and matting | Food segmentation | Performs pixel-level matting on food items in an image and returns the matting result. | 2020-06-30 | China (Shanghai) | |
Clothing segmentation | Performs pixel-level matting on clothing in an input image and returns the matting result. | 2020-06-30 | China (Shanghai) | ||
Image production | HD color transfer | Recolors an HD image while ensuring that the colors of the human portrait remain unchanged. | 2020-06-30 | China (Shanghai) | |
Image color enhancement | Optimally adjusts the saturation, brightness, and skin tone of an input image. | 2020-06-30 | China (Shanghai) | ||
Photo style imitation | Transfers the style of a reference image, such as lighting and color, to a target image without affecting the original structure. | 2020-06-30 | China (Shanghai) | ||
Image recognition | Fruit detection and recognition | Recognizes 60 common types of fruits and 16 types of nuts. | 2020-06-30 | China (Shanghai) | Unpublished |
Image analysis and processing | Chest CT lung nodule detection | Provides auxiliary diagnosis of lung nodules from input DICOM images of conventional chest CT scans (such as a single 5 mm sequence. The API accepts only single sequences). | 2020-06-30 | China (Shanghai) |
May 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Video liveness detection | Detects whether the face in an input video is from a live capture or a recaptured screen. | 2020-05-20 | China (Shanghai) | |
Segmentation and matting | Sky segmentation | Recognizes the sky area in an input image, separates it from the background, and returns the segmented foreground image. | 2020-05-20 | China (Shanghai) | |
Animal segmentation | Recognizes the outline of an animal in an input image, separates it from the background, and returns the segmented foreground animal image. | 2020-05-20 | China (Shanghai) | Unpublished |
April 2020
Category | API Name | Description | Release date | Region | References |
Image production | Image composition aesthetics scoring | Analyzes an input image and outputs an aesthetic score for its composition. | 2020-04-20 | China (Shanghai) | |
Image exposure scoring | Analyzes an input image and outputs a score for its exposure. | 2020-04-20 | China (Shanghai) | ||
Image sharpness scoring | Analyzes an input image and outputs a score for its sharpness. | 2020-04-20 | China (Shanghai) | ||
Video production | General video generation | Intelligently generates short marketing videos from an input source video. | 2020-04-20 | China (Shanghai) |
March 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Create face database | Creates a face database. | 2020-03-20 | China (Shanghai) | |
You can view the database list | Lists face databases. | 2020-03-20 | China (Shanghai) | ||
Add face samples | Adds face entity data to a face database. | 2020-03-20 | China (Shanghai) | ||
Query face samples | Queries face entity data in a face database. | 2020-03-20 | China (Shanghai) | ||
List face entities | Queries the list of face entities in a face database. | 2020-03-20 | China (Shanghai) | ||
Update face entity | Updates face entity data in a face database. | 2020-03-20 | China (Shanghai) | ||
Add face data | Adds face data to a specified database. | 2020-03-20 | China (Shanghai) | ||
Search for faces | Searches a database for similar face images based on an input image. | 2020-03-20 | China (Shanghai) | ||
Delete face | Deletes face image information from a specified database. | 2020-03-20 | China (Shanghai) | ||
Delete face entity | Deletes face entity data from a face database. | 2020-03-20 | China (Shanghai) | ||
Delete database | Deletes a specified face database. | 2020-03-20 | China (Shanghai) | ||
Hand gesture key points | Retrieves information about 21 key points of a hand gesture. | 2020-03-20 | China (Shanghai) | Unpublished | |
Body posture key points | Retrieves information about 18 key points of the human body. | 2020-03-20 | China (Shanghai) | ||
Pedestrian detection | Detects human bodies in an image. | 2020-03-20 | China (Shanghai) | ||
Face restoration and enhancement | Clips, aligns, and enhances the details of a face in an input image, and then merges it back into the original image. | 2020-03-20 | China (Shanghai) | ||
Face filter | Changes the overall style of an image. | 2020-03-20 | China (Shanghai) | Unpublished | |
Face retouching | Retouches faces in an image, including skin smoothing, whitening, and removing dark circles and nasolabial folds. | 2020-03-20 | China (Shanghai) | ||
Face makeup | Simulates makeup application by adding cosmetic elements to further enhance the face retouching effect. | 2020-03-20 | China (Shanghai) | Unpublished | |
Face reshaping | Adjusts the facial contour and features. | 2020-03-20 | China (Shanghai) | Unpublished | |
OCR | Document structure restoration and recognition | Parses the content of an input document and outputs it in a structured format (HTML or JSON). | 2020-03-20 | China (Shanghai) | Unpublished |
Chinese passport recognition | Recognizes key fields on a Chinese passport, including the following: Chinese name (with Pinyin), passport number, passport holder ID, gender, English name, date of birth, place of birth (with Pinyin), nationality, date of issue, date of expiry, issuing authority (with Pinyin), first line of the machine-readable zone (MRZ), second line of the MRZ, and passport type. | 2020-03-20 | China (Shanghai) | Unpublished | |
Food delivery order recognition | Recognizes key fields on a food delivery order, and outputs the store name, phone number, packaging fee, delivery fee, subtotal, other fees, customer discounts, total items, online payment amount, order number, and order time. Currently supports Ele.me delivery orders. | 2020-03-20 | China (Shanghai) | Unpublished | |
Passport MRZ code recognition | Analyzes an input image of a passport's MRZ and outputs 11 pieces of information: type, country code, passport number, name, nationality, date of birth, gender, start date, expiry date, machine check digit 1, and machine check digit 2. This facilitates subsequent information extraction and certificate verification. | 2020-03-20 | China (Shanghai) | Unpublished | |
Product understanding | Home furnishing attribute recognition | Recognizes the style of an input image of a home furnishing model. It currently supports 16 styles, including the following: light luxury, Nordic, retro/nostalgic, other, industrial, Southeast Asian, Ming and Qing classical, Korean, minimalist, Japanese, American country, European/classical, modern, simple European/neoclassical, new Chinese, and Mediterranean. | 2020-03-20 | China (Shanghai) | Unpublished |
Home furnishing SPU recognition | Classifies an input image of a home furnishing model. It supports up to 70 categories. | 2020-03-20 | China (Shanghai) | Unpublished | |
Image recognition | Vehicle model recognition | Recognizes the type of vehicle in an image (full or partial image). Supported categories mainly include sedans, multi-purpose vehicles (MPVs), and SUVs. | 2020-03-20 | China (Shanghai) | Unpublished |
Garbage classification | Classifies the garbage items in an image and provides the specific item names. | 2020-03-20 | China (Shanghai) | ||
Segmentation and matting | Furniture segmentation | Performs pixel-level matting on furniture in an input image. | 2020-03-20 | China (Shanghai) | Unpublished |
Mask refinement segmentation | Refines a coarse mask for an input image and outputs a refined mask. | 2020-03-20 | China (Shanghai) | ||
Image production | Station logo erasing | Erases common logos from an image, such as station logos and Internet platform logos. | 2020-03-20 | China (Shanghai) | |
Image subtitle erasing | Erases standard captions from an image. | 2020-03-20 | China (Shanghai) | ||
Invisible text watermark for images | Adds or extracts a specified text watermark to or from an image. | 2020-03-20 | China (Shanghai) | ||
Invisible image watermark for images | Adds or extracts an image watermark to or from an image. | 2020-03-20 | China (Shanghai) | ||
Visual search | Create database | Creates an image database. | 2020-03-20 | China (Shanghai) | |
List databases | Lists the databases. | 2020-03-20 | China (Shanghai) | ||
Add image data | Adds image data to a specified database. | 2020-03-20 | China (Shanghai) | ||
List image data | Lists the image data in a specified database. | 2020-03-20 | China (Shanghai) | ||
Search for images | Searches a database for similar images based on an input image. | 2020-03-20 | China (Shanghai) | ||
Delete database | Deletes a specified database. | 2020-03-20 | China (Shanghai) | ||
Delete image | Deletes an image from a specified database. | 2020-03-20 | China (Shanghai) | ||
Video understanding | Video thumbnail | Analyzes an input video and outputs multiple video thumbnails. | 2020-03-20 | China (Shanghai) | |
Video shot detection | Splits an input video by shot and returns the split points. | 2020-03-20 | China (Shanghai) | ||
Video production | Video super-resolution | Enlarges an input video to twice its original size. It enhances the video quality by inferring details. The output video is in MP4 format with H.264 encoding. | 2020-03-20 | China (Shanghai) | |
Video color correction | Performs color correction on an input video. It can perform associated color correction based on the similarity between videos. | 2020-03-20 | China (Shanghai) | ||
Video subtitle erasing | Erases standard captions from a video, such as the white captions at the bottom of movies and TV series. | 2020-03-20 | China (Shanghai) | ||
Video logo erasing | Erases common logos from a video, such as station logos and Internet platform logos. | 2020-03-20 | China (Shanghai) | ||
E-commerce video synopsis | Generates a video synopsis of a specified duration from an input video. | 2020-03-20 | China (Shanghai) | Unpublished | |
Movie and TV video synopsis | Extracts a video of a specified duration from an input movie or TV video. | 2020-03-20 | China (Shanghai) | Unpublished | |
Object detection | Object detection | Detects objects in an input image. | 2020-03-20 | China (Shanghai) | |
White background image detection | Detects whether an image has a white background. | 2020-03-20 | China (Shanghai) | ||
Transparent image detection | Detects whether an image has a transparent background. | 2020-03-20 | China (Shanghai) | Unpublished |
February 2020
Category | Interface Name | Description | Release date | Region | References |
Face and body | Expression recognition | Detects and recognizes facial expressions in an image. | 2020-02-28 | China (Shanghai) | |
Body counting | Counts the number of human bodies in an input image. | 2020-02-28 | China (Shanghai) | ||
Face liveness detection | Detects whether a live object (mainly a face) in an image is from a live capture or a recaptured screen. The presence of a face in the image is a prerequisite for liveness detection. | 2020-02-28 | China (Shanghai) | ||
Public Facial Recognition | Recognizes one or more public figures in an image. | 2020-02-28 | China (Shanghai) | ||
OCR | VAT invoice recognition | Recognizes key fields on value-added tax (VAT) invoices (electronic and paper), including the following: check code, reviewer, issuer, invoice code, and payee. | 2020-02-28 | China (Shanghai) | |
QR code recognition | Detects whether an image contains QR codes and outputs the text information (URL or text for each QR code). It supports the recognition of multiple QR codes in an image. | 2020-02-28 | China (Shanghai) | ||
Content Moderation | Text content moderation | Combines behavior and content analysis using multi-dimensional, multi-model, and multi-detection methods to identify spam in text. This helps mitigate risks from content such as pornography, advertisements, spamming, political content, and abuse. | 2020-02-28 | China (Shanghai) | |
Image recognition | Logo recognition | Analyzes a submitted image to recognize the logo information it contains (mainly station logos and trademarks). | 2020-02-28 | China (Shanghai) | Unpublished |
Image segmentation | Facial feature segmentation | Analyzes a frontal face image and performs pixel-level semantic segmentation of the eyes, nose, and mouth. | 2020-02-28 | China (Shanghai) | |
Vehicle segmentation | Performs matting and analysis on the vehicle area in an input image. | 2020-02-28 | China (Shanghai) | Unpublished | |
Image production | Intelligent composition | Performs an aesthetic assessment of an input image and intelligently outputs bounding boxes. You can use these bounding boxes to clip the original image into a better one. | 2020-02-28 | China (Shanghai) | |
Object detection | Vehicle insurance image classification | Classifies input vehicle insurance images. | 2020-02-28 | China (Shanghai) | Unpublished |
Vehicle part recognition | Detects the position and name of vehicle parts in an image. | 2020-02-28 | China (Shanghai) | Unpublished | |
Vehicle damage recognition | Detects the location and type of vehicle damage in an image. | 2020-02-28 | China (Shanghai) | Unpublished | |
Vehicle dashboard recognition | Recognizes information on a vehicle dashboard, such as warning lights. | 2020-02-28 | China (Shanghai) | Unpublished |