UMass Rescue 's repositories
CombinedTechStack
Handle and process large amounts of media data with plug-and-play machine learning models
IntelligentInformationExtractor
A tool to extract important information from audio or text files
ObjectDetection
Object detection using pre-trained model FasterRCNN (trained on MS-COCO) in PyTorch and fine-tune the model for other datasets
ObjectDetectionMicroservice
Object Detection using pre-trained FasterRCNN in PyTorch, trained on MS-COCO dataset.
CanvasTracking
an expo app and backend for canvassing op
FaceDetectMicroservice
Microservice for Face Detection Model
TattooTrace
TattooTrace integrates advanced machine learning models like YOLOv5 and CLIP to analyze and identify tattoos from images and videos. Leveraging the robust backend of Immich, this tool aims to support efforts to combat and understand human trafficking dynamics.
UniversalModelTemplate
A template for creating and testing ML models.
VideoFaceRecognition
This repo contains video face recognition and the client-side demo
596-S22-Backend
Backend Repo for 596RL Spring 2022
Audio-Analyzer-and-Transcription-Tool
End to end information extraction from audio input.
desktop_app
desktop
encoder_decoder
Functions to encode and decode commonly used data formats to support transmission of data over the web
image-similarity
Fine-grained image similarity model based on Deep Ranking.
mobile_app
Ui interface of mobile application. No backend linked.
S2T-Speech-To-Text
An end to end full stack application to transform speech to text and perform further downstream tasks like Text Similarity, Text Summarization, Named Entity Recognition
Weapons_detection_immich
Weapons detection feature based on the immich self-hosted photo and video hosting platform.