havocfixer / mmmesh Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
Hi, thank you so much for your work. Really appreciated :)
I am using IWR6843ISK-ODS with DCA1000EVM to capture .bin file. The antenna pattern in 6834ISK-ODS and 1843BOOST are different and would like your expertise on what changes shall be done in the script to adapt to 6843ISK-ODS for point cloud generation alone.
The antenna pattern and virtual antennas are as shown:
https://drive.google.com/file/d/1bnAqMHeS1rCOPlplNw3qUwXRa7Os5Wmm/view?usp=sharing
Thank you once again.
Hello,
It is a great work! But there is a question about the output dim of CombineModule in your code. It equals 9*6+3+10+1 that is not same as the num 10+144+1+3 =157 in your paper. Why the num of joints is 9? Wish you could answer me.
Thanks!
Hi, is there any way I can access the dataset?
Hi, I am trying to use your code in 1.mmWave_data_generation to capture the real-time data from the AWR1843BOOST radar but every time I try to read data using getFrame() method it gives me readItem as "wait new frame" and itemNum as -2. Can you tell me could be the possible reason for this?
I am using the ubuntu 20.04 subsystem for windows 10 as administrator to run capture.py file.
Thanks!
I appreciate your great work.
In pc_generate.py, you are trying perform FFT along 4 virtual antennas.
elevation_ant = virtual_ant[2 * num_rx:, :]
However, from the TI website (https://e2e.ti.com/support/sensors-group/sensors/f/sensors-forum/1018413/iwr1843boost-range-elevation-heatmap-processing), they suggest perform FFT along two separate rows of virtual antennas.
Does your method follow the same strategy provided by TI?
Thanks for you great work!
The radar type I use is IWR1443. When I run capture.py in step 1-mmWave_data_capture, I can get mmWave radar data successfully and the data is not empty. But it still prints “Packet Lost! Please discard this data.”
How can I deal with this problem?
Hi,
I am very impressed by your work, I am now trying to capture some data with a similar setting as yours. Converting radar ADC data to PC data is very clear, but I wonder how to convert joint coordinates to SMPL pose parameters.
Thank you a lot.
I want to ask the author a question. Nothing happens after the capture.py file is run. My steps are as follows: first connect DCA1000EVM and awr1642, and power on, and finally run capture.py, but the bin file I collected is empty. Do I need to configure the parameters with mmstudio software first, and then run capture.py? I tried but it seems to have failed. Can you give the correct steps to run capture.py? Thanks!
Hi, I am just wondering about the piece of code that generates the x,y,z coordinates from the mmwave frames. Appreciate it if you could point that.
Thank you!
I was wondering if I was missing a step - I'm working with some of the .bin files that I get from the mmWave generation, and I was wondering how it goes from these .bin files -> .dat files. I see in the pc_generation.py, that a Point Cloud is generated, but I was confused on how it gets to the .dat files in the 3.deep_model/data.py. i.e. convert the radar data to the .dat formats (trans.mcl.20p8a) I'm not sure if I missed a step and any help would be appreciated.
I'm confused about the data.py file,
What are the meanings of the data loaded in trans.mcl.20p8a.dat, pmat.mcl.20p8a.dat, betas.mcl.20p8a.dat, and Genes.mcl.20p8a. dat in lines 13 to 17?
And how to get these data?
I'm not sure if I missed a step and any help would be appreciated.
Hello! Thank you for this great work, we are currently using our own radar and motion capture equipment to make a dataset for training, we have the following practical problems:
mmMesh: Towards 3D Real-Time Dynamic Human Mesh Construction Using Millimeter-Wave
是个非常棒的工作 再次感谢您的开源
在对你工作的学习过程中,我有些问题像向您请教:
Can you tell me what energy in pc_generation.py means? Is it reflection intensity? thanks
Thanks for your excellent job! I've read the paper and the code, there are still some questions confuse me. I was wondering is I can use the bin files I get using openradar to skip the step1.mmWave_data_capture, I tried but didn't work. Thus I want to know the point cloud shape stored in /pc.mcl.20p8a.dat file ,so that I could use the point cloud I've got from other method to run the infer_model.py, Thanks a lot!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.