Comments (9)
I have the same question. However when I do x/128. - 1.
on the sample data, the result doesn't match with the given .npy
from kinetics-i3d.
@zekun-li Hi, the provided *.npy file has shape (1, num_frames, 224, 224, 3), I wonder what "1" refers to? And what's its value?
from kinetics-i3d.
@egg347 That dimension corresponds to batch size.
from kinetics-i3d.
@zekun-li Thanks a lot! Have you solved your problem above? It seems that the sample data firstly rescale to [-1,1](videos to videos),and secondly rescale to [0,1](videos to *.npy file).
By the way, I'm confused about the output of the sample. According to the "evalutate_sample.py", it prints Norm of logits、out_predictions[index]、out_logits[index]、kinetics_classes[index])。So what Norm of logits and out_logits[index] can be used for?
With no offense, are you Chinese? Your id looks like a Chinese name. Would you please leave me your EMAIL or your WECHAT id if you don't mind? It comforts me to communicate in Chinese.
from kinetics-i3d.
@zekun-li I also have the same problem on preprocessing. No matter rescaling the image on R, G, B channel individually or across RGB channals, the results all can't match with the given .npy. Have you solved the problem?
from kinetics-i3d.
@zekun-li @egg347 @TianjiPang @seann999 : Hi guys, were you able to resolve the scaling. If yes, would you mind sharing what you did. Thanks
from kinetics-i3d.
I think I simply did what I first guessed, x/128.0-1.0.
from kinetics-i3d.
@seann999: I did the same too, I hope the extracted feature are alright.
from kinetics-i3d.
@zekun-li @seann999 @vivoutlaw @TianjiPang @egg347 I saw many people use x/128.0-1.0
as image normalization. But I still have doubts . For example, if I apply this kind of preprocessing to my frames extracted from v_CricketShot_g04_c01.mp4
, the prediction result I get is robot dancing
. I tried many videos, none of the prediction is correct.
So when you say your result doesn't match with the given .npy
, do you mean the prediction is totally wrong (as in my case), or the prediction is correct (just the softmax score distribution is different)? Thank you very much. Looking forward to your reply.
from kinetics-i3d.
Related Issues (20)
- Optical flow rescaling HOT 5
- How to create a custom action recognition model HOT 2
- Inflating pre-trained models HOT 1
- Trying to create frozen graph, so i can convert it into tflite for android
- Training with different architectures
- customize actions class in the model HOT 1
- offline usage
- rgb.npy and flow.npy HOT 3
- training from scratch HOT 2
- which one is the checkpoint? HOT 5
- The problem of receptive field in I3D paper
- missing videos
- Does the video need to be cropped? HOT 11
- Calculation of TV L1 flow HOT 1
- dependencies issues HOT 1
- Struggling to learn using Opt. Flow HOT 2
- Incompatibility issues
- Run time of I3D on edge decives
- Is there Model File(.pth or .pt) that pretrained with Imagenet+Kinetics?
- I found the pth file which pretrained on Kinetics400 HOT 1
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 kinetics-i3d.