ibrarmalik

How is plenoxels a direct predecessor of gaussian splatting?

By output you mean the extracted surface geometry? Or are you directly rendering NeRFs in VR.

tmux works out of the box for me.

I tried Zellij and couldn’t get the Alt key to work on mac. And then when ssh’ing into a server I couldn’t see some of the icons because it required a special patched font.

I’m assuming this is taking that into account. Otherwise why would it compute a route?

You’re under the right paper for doing this. Instead of one big model, they have several smaller ones for regions in the scene. This way rendering is fast for large scenes.

This is similar to Block-NeRF [0], in their project page they show some videos of what you’re asking.

As for an easy way of doing this, nothing out-of-the-box. You can keep an eye on nerfstudio [1], and if you feel brave you could implement this paper and make a PR!

[0] https://waymo.com/intl/es/research/block-nerf/

[1] https://github.com/nerfstudio-project/nerfstudio

The Luma folks made something similar: https://apps.apple.com/app/luma-flythroughs/id6450376609?l=e...

The new models and data would stay at OpenAI. You can have thousands of researchers and compute, but if you don’t have “it”, you are behind (ask Google).

In Microsoft he still has access to the models, and that’s all he needs to execute his ideas.

Also https://x.org

Same. I like it better than the vscode vim emulation, which is more “strict” and turns to the visual mode when selecting anything with the cursor. Which I personally think is the worse part of vim.

Where does the model come from? Hard to trust the license when we have no idea with what data it has been trained.

I think people are smart enough to receive extra information and do whatever they want with that.

Its a good narrative to have as you prepare for layoffs.

I used to love pacman, but I never used Arch professionally so I can imagine it can have its share of problems.

Or practical vs. philosophical.

If you have a limited number of images of the same scene, with NeRF you can generate new images from different positions and angles (novel view synthesis).

But this only works with rigid scenes: e.g. if you apply NeRF to images of a person, they cannot move between the pictures.

This is what HyperNeRF is trying to solve. If there are pictures of a person, and in one of them they are smiling but on another not, 1. this method will not fail, and 2. looks like it will give reasonable new views/images.

I like Andreas Geiger’s lectures on U. Tübingen [0]. Quite recent, and I think the topics they cover are good.

[0] https://uni-tuebingen.de/fakultaeten/mathematisch-naturwisse...

Indeed building imaginary scenarios helps dissapear any empathy.

This is interesting. The biggest problem I have with my Bose is that they are listening to another device (e.g. my phone) instead of my macbook, even if they are Bluetooth connected. What I would do with the authors app is disconnect my phone from the headphones. I'm not sure I can do this with Toothfairy.

USB-C devices are the kind that you plug and unplug constantly. Unlike the power cord or Ethernet.

You're right! I meant back when I used it.

I believe you are talking about epublibre.org! I have never seen something similar to it in any language I understand. They version control the books, have an "issues" system to alert of errata, and do their own editions. I have even seen books translated by the community.

I stopped downloading spanish books some time ago so I don't know if it still holds up.

For version controlling this could be useful.

If you have the flu you should wear a mask. We are enforcing mask-wearing and doing lockdowns because of COVID asymptomatic infections, these methods don't make much sense for the flu.

People just post pictures of headlines without any source, so they weren't very successful if that was the reason.

Really weird example. So a daycare should hire a pedophile if he wasn't ever convicted?

Pinterest does not develop the Google search engine. If your search results are ruined it's because of Google.

The idea behind Chernoff faces (or using faces for data visualization) seems good: we humans are very good at distinguishing faces, so we can quickly find groups and outliers if the data is encoded with a face.

But we have to be careful with this. Changing facial expressions is not the same as increasing the height of a barplot, we're relating features with expressions and the visualization might express things that you don't want.

There is a very famous example for this in "Life in Los Angeles" (1977) by Eugene Turner [1]. Maybe you can infer the data well but in the end this just ends up being a map of angry black people. The choice of features and how to visualize them is clearly racist.

[1] https://mapdesign.icaci.org/2014/12/mapcarte-353365-life-in-...

I think it was obvious that he doesn't believe that and he's just playing the victim. Nobody in the comments has mentioned anything about white males not having an opinion, and this is a typical argument (and a strawman) that people use to derail the discussion.

Can deconvolution create new data? I thought it was just a way to upscale images.

Maybe they don't know how to create it. You don't need to be a good cook (or even know how to cook) to say that a dish is too salty.