New hands-on tutorial on The Main Thread 🚀

I built a real-time emotion detector in Java using #Quarkus, #JavaCV, #WebSocketsNext, and #LangChain4j.

The pipeline streams webcam frames → detects faces locally → sends only the cropped face to a multimodal LLM (GPT-4o/Gemini) for emotion classification.

High-throughput + low latency = fun and surprisingly practical.

👉 https://www.the-main-thread.com/p/real-time-emotion-detection-quarkus-langchain4j-webcam

#JavaDevelopers #AIPipelines #MultimodalAI #FaceDetection

📸 Enhance Your AI with Precision Face Detection Data!

Supercharge your face detection models with our Face Detection Dataset—curated for accuracy, diversity, and scale. Whether it's for security, or biometrics.

🔑 Key Features:
✅ High-quality images
✅ Diverse demographics
✅ Comprehensive annotations
✅ Ready-to-use for training and testing

👉 Access the dataset here: https://gts.ai/dataset-download/face-detection-dataset/

💡 #FaceDetection #AIDatasets #AITrainingData #ComputerVision #GTS

Face Detection Dataset

Face Detection Dataset comprises 16.7k images and 2 annotation files, each in a distinct format. The first file, labeled "Label,".

FYI - If you're using the #facedetection app for #Nextcloud and running it on #Fedora #Linux, you'll need to recompile the PHP module for #dlib when you upgrade to Fedora 40.

#facedetection
This is pretty neat. Not the huge bloated setup one needs, but a tiny fine minimum.

"This is an open source library for CNN-based face detection in images. The CNN model has been converted to static variables in C source files. The source code does not depend on any other libraries. What you need is just a C++ compiler"

How well it works, I will find out.
I have been playing with dlib's builtin face detection, it was ok but not perfect.

https://github.com/ShiqiYu/libfacedetection

GitHub - ShiqiYu/libfacedetection: An open source library for face detection in images. The face detection speed can reach 1000FPS.

An open source library for face detection in images. The face detection speed can reach 1000FPS. - GitHub - ShiqiYu/libfacedetection: An open source library for face detection in images. The face ...

GitHub
Gestern #calliopemini Online-Session: Motoren und Servos. Die Nacht kein Auge zugemacht🥱. Hier ist das Ergebnis😂. #CalliopeV3,#Esp32Cam und #RaspberryPi5 in einem 🤖-Projekt vereint.#FaceDetection per Stream und ein Roboter der #obstacles ausweicht. #micropython #STEM #robotics

Testing *deface* (https://github.com/ORB-HD/deface), a command line application to detect and anonymize peoples faces in video and images. I tested earlier an image (https://todon.nl/@joenepraat/111703627023452784). This is a video I made after the sit-in inside #UtrechtCS had ended, and the crowd walked to the Stadskantoor (City office) on the other side of the station.

I'm very satisfied with the results. This video is rendered with the default settings, except the audio was kept with the -k option. Best open-source face anonymizer I could find (maybe I missed something). It would be great if someone would make a mobile app based on this code, so that people who film demonstrations could anonymize and upload on the fly.

Edit: I know about #ObscuraCam, but it looks that the development has stalled and it is not as good as this utility.

#FreePalestine #Palestine #CeasefireNOW #FaceDetection

GitHub - ORB-HD/deface: Video anonymization by face detection

Video anonymization by face detection. Contribute to ORB-HD/deface development by creating an account on GitHub.

GitHub
Provision ISR, a SICUREZZA con il meglio delle tecnologie "designed in Israel": Provision ISR invita operatori e prpferssionisti del settore a SICUREZZA 2023 - 15-17 novembre Fiera Milano - per scoprire il meglio della tecnologia per la sicurezza "designed in Israel". DDA Video Analytics di seconda generazione: l’intelligenza artificiale di Provision-ISR. Nuovo...
#Provision-ISR #sicurezza2023 #facedetection #VideoDoorbellWiFi #Intelligenzaartificiale http://dlvr.it/SylPRL #News
Played around with face optimized #yolo would be perfect for an experiment, but 5 landmarks is probably not enough. Too bad. #facedetection

Just finished a joined #webinar with #Synology on the topic of #SurveillanceStation and #DVA3221.

#DeepVideo #FaceDetection #PlateDetection #PeopleCounting

More details in the upcoming articles on https://www.blackvoid.club and https://www.blackvoid.tech

Blackvoid

Synology media partner IT & tech blog focused on Synology NAS, SRM, Docker, and other self-hosted solutions

Blackvoid
I'm currently adding a #facedetection module / API to video4j. The current implementation is using opencv. I'll try a CNN based face detector using dlib next. Should be much faster since it is GPU powered. I want to automatically extract embeddings and use kNN search with #lucene HnswGraph (Hierarchical Navigable Small World graph) to test face recognition / matching.