//=time() ?>
almost a year since i noticed the trend it's accelerating:
training on your own datasets it's not sufficient anymore - even if you dont' copy other you start copying yourself
The third dimension - Ideas/Concepts is even more important
Out-of-Distribution Detection in Multi-Label Datasets using Latent Space of $\beta$-VAE. https://t.co/FYwVtAhwp0
Our article on #RomanLipksi and @annaridler, AI artists using their own paintings and drawings as datasets, is finally up on @FlashArtOnWeb 🤖🎨
Written together with Alex Estorick
Read it here: https://t.co/1P6bUJfMIm
#creativeAI #AIart
Two new @LINZLDS #lidar datasets now avail from OpenTopography: Palmerston North, Manawatu-Whanganui, NZ 2018; Christchurch & Ashley River, Canterbury, NZ 2018 - 1100 sq km of data: https://t.co/Gjfy0DIZiT 📷 1: wind turbines in motion; 📷 2: lidar intensity, Waimakariri River
Multi-organ Segmentation over Partially Labeled Datasets with Multi-scale Feature Abstrac... https://t.co/FafyiGSfX2
I'll point people back to the paper by Nathan Upham and coworkers that just came out last month -- it's the best and newest species-level phylogeny for mammals available, but most of the comparisons are based on a set of 31 genes. Genome-wide datasets are going to shift things!
A Generalizable Method for Automated Quality Control of Functional Neuroimaging Datasets. https://t.co/lDYXoYpO4x
From me for #FridayLiDAR - An update of #lidar and DTM list of open access data! Now you can find 68 various datasets. Additions are from USA, Canada and Brasilia.
#Lidar data extent in @njgin_info - New Jersey
https://t.co/XYIDxdcQhh
ML4A WORKSHOP SIGNUP: In our 2,5-day #workshop with @carolinesinders & @ahandvanish, participants will work to investigate the creation of #subjective #datasets &/ #machinelearning techniques. Accessible for both people with & without technical experience: https://t.co/yPoGOru1RA
Six new @NCALM_UH #lidar datasets now available from OpenTopography. CA sites: Sacramento–San Joaquin River Delta; Horseshoe Lake & Hilton Crk Fault near Mammoth Lakes; Antelope Valley Ranch. Idaho Batholith. Goat Rock Volcano, WA. Campeche, Mexico. https://t.co/zX5InOcfY8
We've just made available 123 @osni_maps Opendata datasets - modern mapping of Northern Ireland available to view in all of our National Library of Scotland Reading Rooms.
https://t.co/RZ3h1z83e4
#NorthernIreland #maps
Finalizing my plots of paleo-topography, this one at the LGM (20000 years before present). Very soon, I hope that my paleo-topography datasets will be available for everyone!
Following
@EnglishHeritage Rob Woodside's talk featuring #Hurst Castle & Spit at #SeaChangeConf yesterday re-found some #NewForest DC datasets showing rollback of spit since 1946. Also Vegetation index mapping of saltmarsh behind spit demonstrating Carbon sequestration potential
Ever wonder 'why' exactly rainbow colormaps are bad? In our EuroVis paper, we provide empirical evidence that they perceptually discretize continuous data. Turns out that the discretization happens in unpredictable ways for different datasets. Sorry 🌈 👎 https://t.co/UuMem45kkh
Adam's newest project, MegaPixels, launches in May. It's an online art project which uses face recognition datasets to tell the history of modern biometrics. https://t.co/Fl871CuKfa
Two #opensource tools developed to reproduce the internal and external anatomy of #fossil #bones (https://t.co/RP7gXne69D). Download some freely available #3D datasets and get started today. #FossilFriday #rstats #Morpho @antonio_profico @zarquon42b
Print texts often depict the ❤️ out of context & oriented “upright” on the apex. But that is not real #cardiac anatomy. Since we often employ 3D datasets to teach & portray complex anatomy, they must be accurate & with in situ orientation. #ClinAnat #MedEd
https://t.co/ttc31TaDSw
Loved the last video @_CodeParade_
Maybe we should make a public repo for furry datasets? I've got around 3500 fursuit / furry pictures with facial feature metadata if it's of any use
Today we're releasing three new ML benchmark datasets, based on handwritten character recognition in ancient Japanese manuscripts.
Kuzushiji-MNIST is a drop-in replacement for MNIST, with larger K-49 and K-Kanji datasets also released
Check it out! https://t.co/3laNPrGM7l ✨