Comments (2)
Thank you for your interest in our work. I think you can find the needed information under https://github.com/zhengshou/scnn/tree/master/experiments/THUMOS14/network_proposal/result/
from scnn.
Thanks!
I checked the both https://github.com/zhengshou/scnn/tree/master/experiments/THUMOS14/network_localization/result
and
https://github.com/zhengshou/scnn/tree/master/experiments/THUMOS14/network_proposal/result
But I'm not sure what exactly the columns do.
The information I have so far is:
save the format to csv following the previous
tmp = strsplit(videoname,'_');
seg_swin(end+1,1) = str2num(tmp{end});
seg_swin(end,2) = window_stride;
seg_swin(end,3) = start_frame;
seg_swin(end,4) = end_frame;
seg_swin(end,5) = start_frame/framerate;
seg_swin(end,6) = end_frame/framerate;
seg_swin(end,12) = 1-win_overlap_rate;
also the proposal confidence seems to be seg_swin(10).
But in the localization file, the score is changed to 0 and new scores on 9, 11 and 12 are shown.
What do they mean?
If I want to compare the temporal proposal performance, which result should I use?
Also, do I need to do NMS? clipping? Posterior prob multiplication based on their length?
I just want to make sure that the result we report is not degraded due to any processing missing.
from scnn.
Related Issues (20)
- demo problem HOT 2
- experiment/THUMOS14/network_proposal/result/res_seg_swin.mat HOT 3
- issue about demo result HOT 1
- Changes to VideoDataLayer break test cases for C3D HOT 1
- extract_image_features.bin HOT 3
- The question about eval result HOT 2
- The problem of the run_demo.m and the demo_list_test_uniform16_proposal.lst HOT 2
- Question of the training details of training my own models
- Cannot use C3D v1.1 for demo HOT 3
- prob files in pred/pro/output/ are not created HOT 5
- the outputs of proposal network (probabilities of proposals) are same and very low HOT 1
- MEXaction2 HOT 1
- run demo problem about 000001.prob
- Why not use cls network in prediction?
- how windows 10 ?
- !@
- Bug in run_demo.py at line 194.
- label tool
- The pretrained model URL has no content
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from scnn.