Viewing profile — potac

Not sure if I understood correctly. Can we as individuals contribute to any of these projects?

Location: Spain/UK Remote: Yes Willing to relocate: Yes (Europe) Technologies: 3D reconstruction, neural rendering, multi-view geometry, Diffusion models, AR/VR, Python, PyTorch, O…

I'm not saying the mayor should change what is not under his radar. I'm saying the mayor should attract private companies for locals that don't want to be public state workers.

Do you know the funny thing? There is a cycle path from Pontevedra that stops ~500 meters before Marin (although still Pontevedra jurisdiction). It has been like this for years. Th…

It's true there is a bus to Marin every 20min but it uses a _single_ fixed route. People that live far away (>1km) from this path spend less time driving to Pontevedra than walking…

Huh! I didn't know this. Thanks for clarifying. Yeah, we all know what the Xunta is doing...

I'm from Pontevedra. It has been the major's long-term project (~ 20 years) to make the city for the pedestrians: and he's done it. This works mainly because of two things: 1) the …

I don't really get why they need the occlusion volume? What is its functionality? Doesn't rendering a pointcloud from a novel viewpoint already give the (dis-)occlusion mask?

Just look at Pontevedra in Spain. 90% pedestrianised city and several years with 0 traffic related deaths. Lovely city!

Thanks. What was confusing me is the kernel size 4. Normally in (2D) convolutions you have (in_channels, out_channels, k, k) for a kxk kernel size. In the example above it the k is…

Can anyone explain how conv works in that graph. You have a tensor of shape [2,4,16] and you convolve with a kernel of shape [4,16,8] and that gives you a [2,8] tensor? How's that …