Pablo García Ruiz

👨‍💻 Computer Vision Researcher | 🌐 Passionate about AI & Machine Learning | 📚 Lifelong Learner | 🎙 Sharing tech insights | #ComputerVision #AIEnthusiast

[Artificial Vision Applications (AVA) group - UCO @ucoava]

📰 Excited to introduce our new dataset: Health & Gait.
🚶‍♀️A video-based resource for gait analysis.
💡 Enables health assessment using AI-driven metrics, from gait speed to anthropometry.
📄 Paper: https://t.co/dfXhbGJvEI
💻Code: https://t.co/6FcG1hmLBm
#AI #HealthTech #AVAgroup

[Artificial Vision Applications (AVA) group - UCO @ucoava]

📰Novel method for Large-scale Indoor Camera Pose Estimation!
💡Effectively position any camera arrangement using fiducial markers.
💻Code, tutorials and an app framework: https://t.co/4vWbi3JbVY
💡 Paper: https://t.co/RByRnpWFbD
#ArUco #ComputerVision #AVAgroup #Robotics #SLAM

[Artificial Vision Applications (AVA) group - UCO @ucoava]

👩‍🎓👨‍🎓 El pasado miércoles, @IsaJimenez_99 y @pabgarudev, presentaron con éxito sus líneas de tesis doctoral en el Congreso Científico de Personal Investigador en Formación de la @Univcordoba
✨Pablo recibió Premio a la Mejor Comunicación de Ingeniería y Arquitectura🏅Enhorabuena!

[Artificial Vision Applications (AVA) group - UCO @ucoava]

📷 @pabgarudev habló sobre Fiducial Objects y sus aplicaciones (estimación de pose de la cámara, calibración, #AR).
🤗 @IsaJimenez_99 habló sobre la Proxémica y cómo detectarla automáticamente en imágenes.

[Carlos Santana @DotCSV]

Ahora sí, con la valoración del público, podemos decir que acabamos el año con un modelo open source (Mixtral-8x7B) equiparable a ChatGPT 3.5, pero que puedes descargar y ejecutar en local.

Eso es 🔥🔥🔥🔥🔥

[Pablo García Ruiz @pabgarudev]

🔍 Proud to present our latest research at @ucoava enhancing camera pose estimation using custom Fiducial Objects, addressing traditional marker limitations. Our work is fully open-sourced.
🔗Project page: [https://t.co/PMCPz8C0BW] #CameraPoseEstimation #FiducialObject

[Carlos Santana @DotCSV]

🔴 ¡PRUEBA MIXTRAL 8x7b FÁCIL!

Todos hablan del nuevo MoE en la ciudad, Mixtral! Rinde muy bien con la ventaja de ser más versátil y claro... open source.

Si queréis probarlo podéis descargarlo, o más sencillo aún, testearlo gratis desde la web de Perplexity. 👀🔥

[Pablo García Ruiz @pabgarudev]

🚀 Discover Pearl: Meta's new RL agent! Modular design meets real-world challenges with smart exploration, safety features & dynamic action spaces.
🔗 Link to paper: https://t.co/aa7DM9fu7s.
#ReinforcementLearning #AI #TechInnovation

[Pablo García Ruiz @pabgarudev]

🚀 "Efficient Large Language Models: A Survey" delves into the high resource demand of LLMs. Wan, Wang, Liu et al. categorize solutions and invite collaboration.

🔗 Find more on https://t.co/50q3WQO5HU.

#AI #Efficiency #TechInnovation #LLMs

[camenduru @camenduru]

HierSpeech++: Zero-shot Speech Synthesis 🎙 Voice-Clone Colab 🥳

Thanks to Sang-Hoon Lee ❤ Ha-Yeong Choi ❤ Seung-Bin Kim ❤ Seong-Whan Lee ❤

🌐page: https://t.co/QCcWcphxQI
📄paper: https://t.co/UxXNPHVGOR
🧬code: https://t.co/1CmZcUaRNi
🦒colab by https://t.co/uUxzMCZb53: please try it 🐣 https://t.co/WY41H6R4xx

[Stability AI @StabilityAI]

Introducing SDXL Turbo: A real-time text-to-image generation model.

SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one.

The code, research paper, and weights for non-commercial use are now available on our website.

You can test SDXL Turbo on Stability AI’s image editing platform @Clipdropapp, with a beta demonstration of the real-time text-to-image generation capabilities.

Learn more: https://t.co/L39rZWf9F7

[Artificial Vision Applications (AVA) group - UCO @ucoava]

"Proxemics" studies how humans use personal space as nonverbal communication. @IsaJimenez_99 et al. propose a new method, ProxemicsNet 🤖, that using #ComputerVision, automatically classifies interactions in images.
🗒️ https://t.co/i8Hakcb0XZ
💻Code: https://t.co/DeLYzfnHGx
#AI