"Large Scale Visual Food Recognition. (arXiv:2103.16107v3 [cs.CV] UPDATED)" — A food dataset with 2,000 categories and over 1 million images that can be used for food recognition.

Paper: http://arxiv.org/abs/2103.16107
Code: https://github.com/Liuyuxinict/prenet/

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
The distributions over each cat…

@BobWilliams I guess it might be possible to do something from the browser end but for that I’ve only explored things like writing a browser-based client in JavaScript 🙂

That’s probably the solution which would cater to most people since it should run on most (any?) platforms but it turned out that there wasn’t an existing pure JavaScript solution that I could leverage. I’d have to write all of it on my own. I got started but didn’t go beyond displaying a timeline since all the work that remained seemed too daunting and it was too long to wait for the payoff.

So I just found the best native (macOS) client that was open source, and modified it to add the features I wanted. But I’ll probably go back to the JavaScript client at some point since it should help a lot more people (possibly) …

@BobWilliams Making changes at the instance end would be one way to set things up for a larger group of people I guess 🙂

For me, personally, there aren’t a lot of Fediverse implementations which have the features I want and the ones that do are implemented in programming languages that I’d have to learn, or the implementation itself is rather cumbersome, as is the case with Mastodon.

So I opted to do the changes at the client-end with specific client apps since that’s a little bit more manageable. That presents certain problems but less so than managing the server-side stuff to do things in an efficient way … at least for me.

Plus, if you get the client working right, you can use it to connect to any instance and so be able to move to a new server if you wanted, etc. 🙂

"Directed Diffusion: Direct Control of Object Placement through Attention Guidance. (arXiv:2302.13153v1 [cs.CV])" — Controlling object placement in diffusion models by way of attention guidance.

Paper: http://arxiv.org/abs/2302.13153

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Directed Diffusion (DD) key res…

@BobWilliams Happy to help in any way I can 🙂 I’ve been running some experiments myself in reducing my home timeline by filtering but only by way of customising my client app since the generally available apps don’t seem to have that feature.

I find lists much more useful now that I can filter out list members from my home timeline but somebody else said something which leads me to believe that other people might be disregarding the home timeline altogether (since they can’t filter by lists membership) and simply putting all of the people they are following into lists so that they only deal with lists.

Not sure if that’ll help you but just mentioning in case it does …

A total of 98 papers in the cs.CV category on arXiv.org today — 35 new, 63 updated.

Gentlepeople, start your engines 🙂

#AI #CV #NewPapers #DeepLearning #MachineLearning

@davemark @RachYall Thank you. Now I’m even less inclined to play it because I get to emotionally invested in stuff…. I just spent days thinking about the end of “Uncharted 4” and that was a happy ending 🙂

@davemark Please lemme know what you think when you do 🙂 I probably won’t get to it this year. I’m slowing down on the gaming and we just got back to playing “Rise of the Tomb Raider” after like 4 or 5 years?

And then “Diablo IV” is coming soon so probably that first …

@BobWilliams Definitely, not just you 🙂 I feel the same way and make a conscious effort to reduce the number of people I follow when my timeline starts becoming too busy …

I hit a point where I’ll feel as if I’m just drowning trying to keep up and that I’m mostly reading but not interacting. Then I try to cut down on the timeline (or move some people over to lists so that they don’t hit the main timeline) so that I have less stuff to deal with. Seems to improve things … at least for me …

@matthras I’m curious about the “too smart” part too. Are you simply saying that ChatGPT is not very smart? Or something that I’m not getting?

@vicki

@davemark Only tangentially related to your question (and no, not an answer 😛) I’ve been debating over playing the game for years now …. Given what I’ve seen written about the game, it sounds as if it’s got a lot of emotional pull, but it also feels to me (partly based on the TV show but I felt the same way before) that it could be emotionally draining — that you’d get to care about the characters but then be devastated at what they go through …

So I stil keep going back and forth about actually playing ,… I have the original (and possibly the new remake) in my library as part of PlayStation Plus (or whatever they are calling it) but I don’t know if I want to play …

Call me contrary, but when I see people saying “I want to work at <Apple/Google/Microsoft>”, I ask, “Why?”

Somebody at work told me recently that Apple would want to hire me because I did/knew something only like 10 people in the world could do and 7 of them worked at Apple. So I was like, “If 7 of them are at Apple, why would Apple want to hire me?”

Plus, I don’t want to work at Apple 😛

I’ve been at Apple as a visitor and seen what the culture is like, and I really don’t want to work at that kind of place. Plus, I really like working from home and based on what I saw (and what I’ve read since the pandemic) Apple really does not seem to like people working from home. Based on what I saw, that tallies.

So why would I work for a company which has diametrically opposed viewpoints to mine? Just because it says “something” about me? Or because I think they might pay me a lot better and money fixes everything?

I do think that if I did that, it would say something about me, but not the same “something” that others are thinking of. Personally, I’d think that I put other people’s opinion of what I am over my own comfort. And I don’t want to do that …

#Work #Companies #Opinions

Boosted 3 papers in the cs.CV category (and one outside) out of a total of 49 new and updated papers on arXiv.org today.

On to other things now …

#AI #CV #NewPapers #DeepLearning #MachineLearning

"In What Languages are Generative Language Models the Most Formal? Analyzing Formality Distribution across Languages" — Measuring the formality of the generated text for different languages using multilingual generative language models.

Paper: https://arxiv.org/abs/2302.12299

#AI #NewPaper #DeepLearning #MachineLearning #Language

<<Find this useful? Please boost so that others can benefit too 🙂>>
Differences between formal and …

"ISS: Image as Stepping Stone for Text-Guided 3D Shape Generation. (arXiv:2209.04145v6 [cs.CV] UPDATED)" — Using 2D images as a stepping stone for creating 3D shapes and eliminating the need for paired text-shape data.

Paper: http://arxiv.org/abs/2209.04145
Code: https://github.com/liuzhengzhe/ISS-Image-as-Stepping-Stone-for-Text-Guided-3D-Shape-Generation

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Our novel “Image as Stepping St…

"Modulating Pretrained Diffusion Models for Multimodal Image Synthesis. (arXiv:2302.12764v1 [cs.CV])" — Multimodal Conditioning Modules (MCM) for enabling conditional image synthesis using pretrained diffusion models so that you can generate images using not just a text prompt, but additional input such as a segmentation map or a sketch.

Paper: http://arxiv.org/abs/2302.12764

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Multimodal conditioning modules…

"Surface Recognition for e-Scooter Using Smartphone IMU Sensor. (arXiv:2302.12720v1 [eess.SP])" — Detecting whether an e-scooter is on a paved road or a sidewalk using the Inertial Measurement Unit (IMU) sensors on a smartphone.

Paper: http://arxiv.org/abs/2302.12720

#AI #CV #NewPaper #DeepLearning #MachineLearning

<<Find this useful? Please boost so that others can benefit too 🙂>>
Left: an example of a street wi…

We (the wife and I) like playing console games — we especially like co-op games that we can both play at once. But we hate split-screen co-op since we get confused as to who is on which part of the screen 😛

So we are kind of limited on the games we can play. Because of that, when we have nothing else to play, we do try single-player games …

Yesterday was one such day where we tried a bunch of single-player games on PS5 to see if we can find something that we liked.

First up was, “Outriders” — the graphics looked great, it was a science-fiction storyline and so we thought we might like it. Unfortunately, it’s a mostly “shooting” game. You have to keep shooting people to advance and we don’t like that kind of game much …

Sure, we don’t mind if there’s some combat, but we prefer a storyline, puzzles, exploration, that kind of thing …

So, we tried “ReadySet Heroes” next. This one was co-op, but split screen. Given that we didn’t have a lot of choices, we decided to give it a try. But unfortunately, it was just dungeon crawling. No story, nothing of interest unless you just like mindlessly bashing things. Next!

The last one we tried was “Omno” and here we hit the Goldilocks zone … almost 😛 It was single-player, but it wasn’t frantic shooting/button mashing. It had a gentle, explore-at-your-pace kind of gameplay and lots of puzzles. Sure, we had to pass the controller back and forth between us but we still enjoyed it way more than any of the others and that’s the one we stuck with for the rest of the day 🙂

#Gaming #PS5 #CoOpPlay

A total of 49 papers in the cs.CV category on arXiv.org today — 34 new, 15 updated.

#AI #CV #NewPapers #DeepLearning #MachineLearning

One of the best consequences of my switch to #mastodon has been exposure to people with hobbies and interests different from mine. By following other instances or particular hashtags I get to learn about and learn from people and communities I would otherwise not have encountered.