interdigitalinc / festa Goto Github PK
View Code? Open in Web Editor NEWSource code for the paper "FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds"
License: Other
Source code for the paper "FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds"
License: Other
Hi. Thanks for sharing your code.
I run FESTA on my GTX 1080TI and change epoch from 500 to 100.
After training, i run the test.sh, but there is the error i cannot solve:
Traceback (most recent call last):
File "/root/.pycharm_helpers/pydev/pydevd.py", line 1483, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
File "/root/.pycharm_helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "/root/FESTA/train.py", line 467, in <module>
main()
File "/root/FESTA/train.py", line 430, in main
net.load_state_dict(checkpoint['state_dict'])
KeyError: 'state_dict'
I have debugged this code, i found there is no dict which named 'state_dict' in model.best.t7.
I am expecting your kindly Reply.
Hi, how is the point cloud visualization in the paper done, it looks very 3D stereoscopic.
Thank you so much for sharing you code ..
May I ask you please about the difference between the result on your paper and the previous works like PointPWC-Net, HPLFlowNet, PV-RAFT ? and how do get a different results for these works while they are not mentioned in their original papers? Is there a specific way to calculate these values?
Thanks in advance and hope you can see and answer my question..
Hi,
Thanks for this wonderful project! Can you share the pre-trained model on Flyingthings3D?
Besides, I only use the geometry information to train FESTA on FlyingThings3D, it cannot achieve the EPE of 0.1253.
Final test results on FlyingThings3D:
mean test loss: 0.024688
mean test epe_3d: 0.134604
mean test acc_3d: 0.363211
mean test 3d_2_sum: 0.686857
mean test mask_sum: 0.160795
mean test epe_3d end: 0.133733
mean test acc_3d end: 0.363290
mean test 3d_2_sum end: 0.687539
mean test mask_sum end: 0.160651
Final test results on KITTI without fine tuning:
mean test loss: 0.028747
mean test epe_3d: 0.153715
mean test acc_3d: 0.197930
mean test 3d_2_sum: 0.579847
mean test mask_sum: 0.071703
mean test epe_3d end: 0.152873
mean test acc_3d end: 0.197633
mean test 3d_2_sum end: 0.580544
mean test mask_sum end: 0.070945.
The results are worse than the results listed in Table2. I wonder if some parameters need to be adjusted!
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.