Guillaume Demarcq

CEO & cofounder at Ikomia -

[Ikomia Imaginarium @ikomiaginarium]

🚀 We built Ikomia Imaginarium, a generative AI SaaS platform! From tech stack to AI deployment, it's all here. Check it out! 👇

https://t.co/LLHKhLCzWh

#AI #SaaS #GenerativeAI #Developers #Entrepreneurs

[Ikomia Imaginarium @ikomiaginarium]

🚀 Big news! Imaginarium is now LIVE! 🎨✨

Create stunning visuals from text or generate unique AI avatars.

🎉 30% OFF for the first 100 subscribers! Don’t wait – unleash your creativity now.

#Imaginarium #GenerativeAI #AIArt #Creativity #Innovation

[Guillaume Demarcq @g_demarcq]

I just published Best AI Diffusion Models: A Comprehensive Comparison and Guide [2024] https://t.co/ITyQ8iT2Jl

[Ikomia @IkomiaOfficial]

New blog post on YOLACT, real-time instance segmentation ideal for autonomous vehicles, medical imaging, and precision agriculture https://t.co/4mtcwYfZFU

- Remarkable real-time performance.
- High accuracy levels.
- Reduced computational load.

#ComputerVision #AI

[Ikomia @IkomiaOfficial]

🚀 Master YOLOR for object detection with just a few Python code lines! 🤖

Discover why YOLOR is reshaping object detection and how you can leverage its capabilities in your projects with minimal coding effort https://t.co/HZIF89uNGr

#YOLOR #ObjectDetection #ComputerVision #AI

[Ikomia @IkomiaOfficial]

ByteTrack multi-object tracking algorithm has been added to Ikomia HUB! https://t.co/6wi3Gp8TRa

You can now easily test it and mix it in a workflow with other open-source SOTA algorithms from the library.

#computervision #opensource #objecttracking

[Ikomia @IkomiaOfficial]

Introducing Onnx Donut, the Python package for exporting a Donut model from Pytorch to ONNX, and running it with onnxruntime 🍩🐍.
Once a model is converted in ONNX, it is automatically quantized to reduce its size on disk.
https://t.co/PANT1REwHS

#AI #Python #pytorch #onnx

[Ikomia @IkomiaOfficial]

📅 Meet us next week at #apidays Paris!

👥 Connect with the Ikomia team at booth 54 to delve into AI APIs and experiment with Computer Vision case studies using our open-source API.

Haven’t booked your ticket yet? Be our guest: free regular ticket https://t.co/LxCGDArCsS

[Ikomia @IkomiaOfficial]

🤔 Facing challenges with Detectron2?

🔧 From setting up your Python environment to handling tricky installation steps, we've covered it all.

👉 Full post here: https://t.co/eI2xHltO8R

#Detectron2 #InstanceSegmentation #AI #ComputerVision #Python

[Ikomia @IkomiaOfficial]

We’ve put together a new case study on SwinIR image restoration 🖼️.

Key insights on mastering SwinIR and integrating it smoothly into your own image processing workflows.

Feel free to share your experiments with us!

#deeplearning #AI #superresolution

https://t.co/QrpB9JquB7

[Ikomia @IkomiaOfficial]

MODNet portrait matting inference algorithm has been added to Ikomia HUB!

https://t.co/KAJJOitLpo

You can now easily test it and mix it in a workflow with other open-source SOTA algorithms from the library.

#computervision #opensource #imagematting

[Guillaume Demarcq @g_demarcq]

@ylecun @huggingface @Scaleway @Meta Hi @ylecun @ClementDelangue , at @IkomiaOfficial we are a French startup 🤗🇫🇷 and we make open source tools for Computer Vision. Nice to see this announce, can we meet and discuss?
(Qui tente rien n’a rien, avec un peu de chance, l’un de vous verra ce message ! 🤓)

[Ikomia @IkomiaOfficial]

Discover how Ikomia bridges the gap between diverse dataset formats, making it easy to load, connect and fine-tune your #computervision models: https://t.co/Xzxq9HFszf .

#AI #opensource

[Ikomia @IkomiaOfficial]

SwinIR Super Resolution algorithm has been added to Ikomia HUB!

https://t.co/uKCXB63Qb8

You can now easily test, modify and mix it in a workflow with other open-source SOTA algorithms from the library.

#computervision #opensource #superresolution

[Ikomia @IkomiaOfficial]

Create a workflow for pose estimation with MMPose with a few lines of #python code https://t.co/of7aRb3Iph.

OpenMMLab projects have been integrated into the Ikomia HUB, simplifying the installation and making it easy to incorporate into your workflows.

#AI #computervision

[Ikomia @IkomiaOfficial]

Text extracting from ID cards using deep learning 🧠

In our latest post, we detail our approach from a recent POC using 5 steps:
- instance segmentation
- cropping
- text detection
- key info extraction
- text extraction

Details & resources: https://t.co/L4ioBpwNVN

#python #AI

[Ikomia @IkomiaOfficial]

👩‍💻 Calling AI Engineers! Tired of compatibility / dependencies issues in vision projects?

🚀 Craft projects faster, join SCALE Beta, deploy easily, and enjoy:
✨ 2 months FREE serverless CPU deployment
✨ 30% LIFETIME discount
Sign up https://t.co/QZd3cYuLcc

#AI #ComputerVision

[Ikomia @IkomiaOfficial]

YOLOv7 Keypoints algorithm has been added to Ikomia HUB!
https://t.co/Rsc9PP9ysd

You can now easily test, train and mix it in a workflow.

#computervision #opensource #keypointdetection #deeplearning #vision

[Ikomia @IkomiaOfficial]

Open source Computer Vision project of interest:

UniControl unifies various image-to-image tasks into a single diffusion model, handling 9 pre-training tasks, including detection, segmentation, and keypoints: https://t.co/mq3Ws2zGpm.

#ComputerVision #AI #opensource

[Ikomia @IkomiaOfficial]

Open source Computer Vision project of interest:

Stable diffusion XL base 1.0, the latest image generation model with more photorealistic outputs and more detailed imagery and composition: https://t.co/Q0x5ra5off
@StabilityAI
#ComputerVision #AI #opensource #stablediffusion