Based on multiple memories of watching the sunset on the Pacific from trips to North America's west coast. I don't know about the rest of you but I'm not a morning person so I love seeing the sunset in the ocean more than rise over it. So these are fond memories.

#MastoArt #FediArt #ArtMatters #AYearForArt #Sunset #DigitalArt #Painting #Abstract #Colorful #MarkOnArt

Find it here: https://mark-tisdale.pixels.com/featured/sunset-dreams-on-the-pacific-mark-tisdale.html

The Milky Way over Snow Capped Himalayas
Credits: Tomas, Havel
#nature #space #astrophotography

The older I get, the more I regret all the people l've lost over the years.
Maybe being a trail guide wasn't such a great idea after all.

Breaking hydroponic cherry tomato news.

Tower of babel made of pizza

#aiart #midjourney

That’s it for another bout of paper reading — 6 papers from the cs.CV category boosted today out of a total of 100 new and updated papers on arXiv.org.

#AI #CV #NewPapers #DeepLearning #MachineLearning

"Composer: Creative and Controllable Image Synthesis with Composable Conditions. (arXiv:2302.09778v2 [cs.CV] UPDATED)" — A way to flexibly control the output image from diffusion models to modify the layout or style of the final image.

Paper: http://arxiv.org/abs/2302.09778

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Concept of compositional image …

"Entity-Level Text-Guided Image Manipulation. (arXiv:2302.11383v1 [cs.CV])" — A more accurate/efficient text-guided image editing/manipulation?

Paper: http://arxiv.org/abs/2302.11383

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Comparison between our method a…

"Open-domain Visual Entity Recognition: Towards Recognizing Millions of Wikipedia Entities. (arXiv:2302.11154v1 [cs.CV])" — Creating a non-task/domain specific, general visual recognition model.

Paper: http://arxiv.org/abs/2302.11154

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
An illustration of the proposed…

"Teachable Reality: Prototyping Tangible Augmented Reality with Everyday Objects by Leveraging Interactive Machine Teaching. (arXiv:2302.11046v1 [cs.HC])" — An Augmented Reality (AR) prototyping tool for creating interactive AR applications using everyday objects.

Paper: http://arxiv.org/abs/2302.11046

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Teachable Reality is an augment…

"'The Taurus': Cattle Breeds & Diseases Identification Mobile Application using Machine Learning. (arXiv:2302.10920v1 [cs.LG])" — A cross-platform mobile application to identify cattle breeds, easily analyze and identify the diseases which cattle suffer from, and to provide solutions the identified diseases.

Paper: http://arxiv.org/abs/2302.10920

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Mobile app screens showing the …

"Fair Diffusion: Instructing Text-to-Image Generation Models on Fairness. (arXiv:2302.10893v1 [cs.LG])" — Reducing bias in generative text-to-image models based on instructions.

Paper: http://arxiv.org/abs/2302.10893
Code: https://github.com/ml-research/Fair-Diffusion

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Stable Diffusion (top row) runs…

A total of 100 papers in the cs.CV category on arXiv.org today — 71 new, 29 updated.

Let’s do some reading 🙂

#AI #CV #NewPapers #DeepLearning #MachineLearning

We need fewer "opinion" columnists and more investigative journalists. Or heck, just journalists who report facts in as unbiased a manner as possible.

I think one way we move toward that is by throwing money at the types of journalism that have value. I'll throw it open here: if I have money to put toward a subscription, where would that be best spent?

(edited for clarity and grammar)

Fighting for a healthy Internet for 20+ years, @mozilla is looking for open source professionals to fill technical and non-technical positions. Check out the jobs now on #OSJobHub

What a difference — I wake up today to 50 posts in my filtered home timeline on Tusker vs. 380 posts in my unfiltered Akkoma web interface home timeline 🙂

Well, this certainly is working. But I think I need a few more tweaks — for one, filtering out hashtags I’m following from the home timeline and for another, possibly showing filtered items as compact cards (like what most clients allows you to do for filtered keywords) instead of simply removing the item altogether.

Hopefully, I’m done at that point 😛

#FediClient #Filtering #Timeline

Now that I’ve got actually functional lists (not ones which duplicate content on the home timeline), I am going a bit crazy creating lists 😛

Originally, I thought I’d only have one list for people who generate a lot of content. But then I realized that I can actually create lists for people who create different types of content (mostly) — a jokes list for humour; an arts list for photographers, artists, and so on …

This is getting a lot more complicated but also a lot more organised 🙂

#FediClient #Lists #Organizing

The first five days after the weekend are the hardest.

No matter how still you’ve been, he’s…

How do you think the unthinkable?

With an ithberg.