Comments (11)
Hello, do you have any progress now? I have the same need. Could we discuss it?
from kinetics-i3d.
from kinetics-i3d.
from kinetics-i3d.
In principle the model should handle any resolution since it is convolutional. Joao
…
On Sat, 28 Jan 2023, 08:17 Coolnerdn, @.> wrote: Hello, do you have any progress now? I have the same need. Could we discuss it? — Reply to this email directly, view it on GitHub <#119 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADXKU2UZYJUGAFBWDQWCX3DWUTIZFANCNFSM5KSWB4KA . You are receiving this because you are subscribed to this thread.Message ID: @.>
Thank you for the response. I want to use my own dataset to fine-tune the I3D network. Is there any requirement on the number of frames? Must it be the number of frames during training?
from kinetics-i3d.
from kinetics-i3d.
from kinetics-i3d.
from kinetics-i3d.
it should be possible to feed an arbitrary number of frames to the network, what is the error you get ? You should just have to replace the 64 constant in places like this by your number of frames: inp = tf.placeholder(tf.float32, [None, 64, IMAGE_SIZE, IMAGE_SIZE, 3]) Joao
…
On Sun, Jan 29, 2023 at 8:37 AM Coolnerdn @.> wrote: I see. There is one more question. Did you use the same number of frames when you train the network? The video lengths I use to fine-tune are not the same, so the number of frames are different and can't be fed directly into the network. How did you solve this problem? Or did you crop the video to the same length during training? 哈哈 @. ------------------ 原始邮件 ------------------ 发件人: "deepmind/kinetics-i3d" @.>; 发送时间: 2023年1月29日(星期天) 下午2:22 @.>; @.@.>; 主题: Re: [deepmind/kinetics-i3d] Does the video need to be cropped? (Issue #119) No, you can use a different number of frames. It may be that it will not work as well if you use fewer frames, but you'd have to try to be sure. Best, Joao On Sun, Jan 29, 2023 at 4:41 AM Coolnerdn @.***> wrote: > In principle the model should handle any resolution since it is > convolutional. Joao > … <#m-454727744888391567> > On Sat, 28 Jan 2023, 08:17 Coolnerdn, @.> wrote: Hello, do you have > any progress now? I have the same need. Could we discuss it? — Reply to > this email directly, view it on GitHub <#119 (comment) > < #119 (comment)>>, > or unsubscribe > https://github.com/notifications/unsubscribe-auth/ADXKU2UZYJUGAFBWDQWCX3DWUTIZFANCNFSM5KSWB4KA > < https://github.com/notifications/unsubscribe-auth/ADXKU2UZYJUGAFBWDQWCX3DWUTIZFANCNFSM5KSWB4KA> > . You are receiving this because you are subscribed to this thread.Message > ID: @.> > > Thank you for the response. I want to use my own dataset to fine-tune the > I3D network. Is there any requirement on the number of frames? Must it be > the number of frames during training? > > — > Reply to this email directly, view it on GitHub > < #119 (comment)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/ADXKU2RISFEA5JAWPB3RQHTWUXYHBANCNFSM5KSWB4KA> > . > You are receiving this because you commented.Message ID: > @.> > — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.> — Reply to this email directly, view it on GitHub <#119 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADXKU2S3FZ4O6QXZWXQZZEDWUYT2ZANCNFSM5KSWB4KA . You are receiving this because you commented.Message ID: @.***>
I mean, I'm not sure how many frames I'm going to input, so I can't replace 64 with a specific number, because the number of frames could be any integer from 20 to 90. Can this data be used directly to train networks?
from kinetics-i3d.
from kinetics-i3d.
from kinetics-i3d.
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
- 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.