All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Amazon Rekognition Custom Labels model support
- Amazon Rekognition Segment API support
- Amazon Rekognition Text detection for video
- Amazon Transcribe Custom Vocabulary and Custom Language Model support
- Amazon Comprehend Custom Entity Recognizer support
- Support audio and document files
- Frame based analysis
- Backlog queuing mangagement system
- Rewrite backend state machines to use Nested state machine and Map state to simplify workflow
- Enable AWS-XRAY tracing on all AWS Lambda functions
- Updated search engine to Amazon OpenSearch 1.0
- Redesigned UI/UX to enable different type of collections
- Advanced search feature to pinpoint timestamps of the search results
- Removed Amazon SageMaker Ground Truth Labeling workflow for face indexing
- added additional Comprehend and Transcribe languages support to Media2Cloud
- workaround audio analysis issue when Comprehend result contains leading punctuation characters.
- Backend: new analysis engine with analysis-monitor, audio-analysis, video-analysis, image-analysis state machines
- Backend: Image support (JPG, PNG, BMP, and selective RAW images), ability to extract EXIF, GPS, XMP information
- Backend: enhance analysis detection
- Backend: server side md5 checksum
- Backend: auto restore object from GLACIER, DEEP_ARCHIVE storage
- AIML: Celebrity detection
- AIML: Face Matching with your own face collection
- AIML: Label detection
- AIML: Moderation detection
- AIML: Person pathing
- AIML: Text detection on image
- AIML: Transcription (subtitle) generation
- AIML: Entity detection
- AIML: Keyphrase detection
- AIML: Sentiment detection
- AIML: Topic detection
- UI: support cropping image and index faces to Amazon Rekognition Face Collection
- UI: Ability to index face and store to your own face collection
- UI: Ability to adjust the confidence level of the detection
- UI: Ability to enable and disable specific detection
- UI: Ability to set language for Speech to Text and NLP processing
- UI: AWS SageMaker Ground Truth integration allows you to create and manage your private work team and crowd source your labeling tasks
- UI: Ability to set pagination to limit each query
- UI: Parse JSON definition file and can ingest multiple files defined in the JSON file
- UI: Player now shows markers of labels being detected
- ES: Enhanced Elasticsearch engine, index all mediainfo, exif, and AIML metadata that allows you to do deep search
- API: Simple APIs to ingest, analyze, and search content
- replace Media Analysis solution analysis engine with new analysis engine
- initial checkin