⚡️ Excited to announce that CoLIE has been accepted to #ECCV2024!
CoLIE uses a neural implicit representation to enhance low-light images according to the Retinex theory.
@p_tingying@ja_schnabel@eccvconf
Paper: https://t.co/EHwJEJpavg
Code: https://t.co/K2ioykLMKL
Excited to share TITAN, a multimodal foundation model for pathology trained on >330k WSIs, synthetic captions, and reports🔬⚡
co-lead w @TongDing99@GreatAndrew90@richardjchen
Thanks to @AI4Pathology@p_tingying!
Preprint: https://t.co/LpgmNlbvMa
Model: https://t.co/gYJxLcjKiS
1/ 🧵Introducing UNICORN,🦄 the first #AI model for integrating multi-stain #histopathology images for classification, designed to tackle missing data during both training and inference. Great colab between @HelmholtzMunich & @TU_Muenchen German Heart Centre! the architecture ⬇️
CoLIE can also be applied on a frame-by-frame basis to restore low-light videos and enhance their visual quality, providing substantial enhancements in clarity and detail across the entire video sequence while avoiding undesired overexposure.
⚡️ Excited to announce that CoLIE has been accepted to #ECCV2024!
CoLIE uses a neural implicit representation to enhance low-light images according to the Retinex theory.
@p_tingying@ja_schnabel@eccvconf
Paper: https://t.co/EHwJEJpavg
Code: https://t.co/K2ioykLMKL
I’ll be presenting our method for volumetric microscopy image deconvolution today at #MICCAI23.
I look forward to discussing approaches for image reconstruction with anyone that stops by.
@MICCAI_Society@p_tingying@huiskenlab
Thanks to everyone who joined my presentation yesterday of our work on zero-shot image restoration at the bioimage computing workshop at @ICCVConference 🇫🇷.
I had a wonderful time at @UniofOxford last week, particularly enjoying the opportunity to connect with researchers who share a common focus on inverse problems within the biomedical field. #OxML23
@nickfloats The neck and collar bones don’t look as natural as they could be. Also the eyes seem a bit independent from each other. But quite hard to tell at a glance.
@nickfloats When generating sci-fi screengrabs, getting the helmet illumination was nearly impossible for me with just text-based prompts. Nicely done.
https://t.co/bs8X0NRrCE