duckworthd

10 ms·

31.
by duckworthd 3y ago
Thank you!
32.
by duckworthd 3y ago
Check out our explainer video for answers to this question and more! https://www.youtube.com/watch?v=zhO8iUBpnCc
33.
by duckworthd 3y ago
It would be my dream to make capturing 3D memories as easy and natural as taking a 2D photos with your smartphone today. Someday!
34.
by duckworthd 3y ago
> This is __really__ stunning work Thank you :) > Is it a consequence of input photo quality, or something else? It's more a consequence of spatial resolution: the bigger the space, the more voxels you need to maintain a fixed re
35.
by duckworthd 3y ago
Not at the moment but an intrepid hacker could surely extend our JavaScript code and put something together. UPDATE: The code for our web viewer is here: https://github.com/smerf-3d/smerf-3d.github.io/blob/mai
36.
by duckworthd 3y ago
Right-o. The web viewer is swapping assets in and out of memory as the user explores the scene. The Network and disc requirements are high but memory usage is low.
37.
by duckworthd 3y ago
Thank you :)
38.
by duckworthd 3y ago
In principle, there's no reason you can't fit multiple City blocks at the same time with Instant NGP on a regular desktop. The challenge is in estimating the camera and lens parameters over such a large space. I expect such a reco
39.
by duckworthd 3y ago
> Are there opportunities, where they exist, to not use optimization or tuning methods for reconstructing a model of a scene? If you know a way, let me know! Every system I'm aware of involves optimization in one way or another, fro
40.
by duckworthd 3y ago
;)
41.
by duckworthd 3y ago
I can't predict the future, but I imagine soon: all of the tools are there. The reason we didn't develop for VR is actually simpler than you'd think: we just don't have the developer time! At the end of the day, only a h
42.
by duckworthd 3y ago
Glad you liked our work! 1) Around 100-150 if memory serves. This scene is part of the mip-NeRF 360 benchmark, which you can download from the corresponding project website: https://jonbarron.info/mipnerf360/ 2) Betwee
43.
by duckworthd 3y ago
Pardon our dust: "images" is a bad name for what's being loaded. Past versions of this approach (MERF) stored feature vectors in PNG images. We replace them with binary arrays. Unfortunately, all such arrays need to be loaded
44.
by duckworthd 3y ago
The pretrained models are already available online! Check out the "demo" section of the website. Your browser is fetching the model when you run the demo.
45.
by duckworthd 3y ago
Thank you :). I'm glad to hear it! Which model are you using?
46.
by duckworthd 3y ago
Gaussian Splatting is heavily inspired by work in radiance fields (or NeRF) models. They use much of the same technology!
47.
SMERF: Streamable Memory Efficient Radiance Fields(smerf-3d.github.io)
630 points by duckworthd 3y ago|143 comments
48.
by duckworthd 6y ago
We did not aim to speed this part of NeRF up. Check out Neural Sparse Voxel Fields ( https://arxiv.org/abs/2007.11571 ) for some effort in that direction. It's 10x faster, but there's still another 10x to go ti
49.
by duckworthd 6y ago
A good question! And a problem yet to be solved!
50.
by duckworthd 6y ago
The work we build off of, NeRF, does. While there's nothing preventing NeRF-W from also representing reflections, we find it captures a more matte picture of the object.
51.
by duckworthd 6y ago
Yes, the images need to be posed. We use COLMAP to obtain camera pose.
52.
by duckworthd 6y ago
NeRF's (and all of photogrammetry's) bread and butter is 3D consistency -- that is, seeing the same object from multiple angles. A 360 degree photo from a fixed position just won't do. As to how to select the best camera an
53.
by duckworthd 6y ago
NeRF was first published in March! Give us time :)
54.
by duckworthd 6y ago
We have depth! Check out our depth video: https://youtu.be/yPKIxoN2Vf0?t=146
55.
by duckworthd 6y ago
Another related project by our friends in NYC: https://twitter.com/Jimantha/status/1289184432553734144
56.
by duckworthd 6y ago
For that, you'll need to convert the representation we have (volumetric radiance field) to on your 3D printer can understand (a mesh?). The NeRF authors use the marching cubes algorithm to do just that. Check out their website: https:
57.
by duckworthd 6y ago
Answered most your other questions below in another comment. > and what are z(t) r(t) in equation 5,6? r(t) is a position in 3D space along a camera ray of the form, r(t) = origin + t * direction. z(t) is the output of our first MLP. Thi
58.
by duckworthd 6y ago
Our code is not released yet. If your data is captured without occluders and in RAW format, you don't need the enhancements we propose :). NeRF can do amazing things with clean data!
59.
by duckworthd 6y ago
It is provided by the dataset we use, but given a new dataset, you can use off-the-shelf tools to obtain it yourself! Check out COLMAP, it's super duper cool: https://colmap.github.io/ .
60.
by duckworthd 6y ago
> so why does it use multi-layer perceptron? is it the same as ANN? why not calling it ANN? Does it have activation? According to Wikipedia, "A multilayer perceptron (MLP) is a class of feedforward artificial neural network (ANN).&q

More ›