Comments (10)
poweryin
commented on September 24, 2024
7
@leviswind,Hi，can you train the i3d on ucf101 successfully? I want to use i3d on ucf101, How can I use i3d model to fine-tune on ucf101? where is the train code?
can you give me some advice.
Thanks
Best wishes
from kinetics-i3d.
joaoluiscarreira
commented on September 24, 2024
Hi,
i think 50-60% accuracy is to be expected when training I3D from scratch on
RGB in UCF101. If you do the same on flow it should get ~80%.
When averaging both we got 88% in the last version of the quo vadis paper.
In summary, i think your training setup should be fine.
Best,
Joao
…On Wed, Jan 16, 2019 at 3:50 AM leviswind ***@***.***> wrote:
We try to train i3d model on ucf101 from scratch, but it converges much
slower with a final validation accuray around 60%. Can you offer some
suggestions on train i3d model without imagenet pretrained model.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#46>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AO6qaj792doQPF6DLeSuo3yYIm6AKfcOks5vDqGNgaJpZM4aCQxU>
.
from kinetics-i3d.
leviswind
commented on September 24, 2024
How to train i3d with optical-flow with imagenet pretrained model? Can you offer some details of training on UCF101 dataset.
from kinetics-i3d.
leviswind
commented on September 24, 2024
Also, what's the convergence speed should be when training optical-flow compared with rgb with imagenet pretraining @joaoluiscarreira
from kinetics-i3d.
joaoluiscarreira
commented on September 24, 2024
The way we did it was that we inflated the weights of the imagenet model
into 3D, then trained the model normally from there, without freezing batch
norm. I think you can find code online for training the model if you search
on google. I tend to remember that the flow model converges faster but this
was a long time ago.
Best,
Joao
…On Thu, Jan 17, 2019 at 7:36 AM leviswind ***@***.***> wrote:
Also, what's the convergence speed should be when training optical-flow
compared with rgb with imagenet pretraining
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#46 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AO6qameRG2etXHL7UwUIC5vL1isSGHVHks5vECgCgaJpZM4aCQxU>
.
from kinetics-i3d.
leviswind
commented on September 24, 2024
However, the input channel of the first conv layer is 2 for flow data compared with 3 for rgb. How to deal with the difference? @joaoluiscarreira . I'm really appreciated for your help.
from kinetics-i3d.
joaoluiscarreira
commented on September 24, 2024
I think we just discarded the weights for one of the input channels in that
first layer, before inflating.
Best,
Joao
…On Thu, Jan 17, 2019 at 11:55 PM leviswind ***@***.***> wrote:
However, the input channel of the first conv layer is 2 for flow data
compared with 3 for rgb. How to deal with the difference?
@joaoluiscarreira <https://github.com/joaoluiscarreira> . I'm really
appreciated for your help.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#46 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AO6qau9xAzRpEZOppvHgaG3JODy5j3pgks5vEQ2KgaJpZM4aCQxU>
.
from kinetics-i3d.
joaoluiscarreira
commented on September 24, 2024
Actually i went back to check what exactly we did and for that particular
layer we average the original weights for the 3 input channels then copied
it twice -- so the initial weights are the same for both flow input
dimensions. But i think it did not make much of a difference compared to
the other option.
Joao
On Fri, Jan 18, 2019 at 9:26 AM João Carreira <[email protected]>
wrote:
… I think we just discarded the weights for one of the input channels in
that first layer, before inflating.
Best,
Joao
On Thu, Jan 17, 2019 at 11:55 PM leviswind ***@***.***>
wrote:
> However, the input channel of the first conv layer is 2 for flow data
> compared with 3 for rgb. How to deal with the difference?
> @joaoluiscarreira <https://github.com/joaoluiscarreira> . I'm really
> appreciated for your help.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <#46 (comment)>,
> or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AO6qau9xAzRpEZOppvHgaG3JODy5j3pgks5vEQ2KgaJpZM4aCQxU>
> .
>
from kinetics-i3d.
leviswind
commented on September 24, 2024
We suffered great overfitting when training i3d with optical-flow similar as training rgb data without imagenet pretraining. The test accuracy is only 50%. Have you met such problems? @joaoluiscarreira
from kinetics-i3d.
joaoluiscarreira
commented on September 24, 2024
As mentioned earlier in the thread, training from scratch on flow got close
to 80%. You could try testing with batch statistics to see if there's some
batch norm moving average problem.
…On Sat, Jan 19, 2019 at 5:22 AM leviswind ***@***.***> wrote:
We suffered great overfitting when training i3d with optical-flow similar
as training rgb data without imagenet pretraining. The test accuracy is
only 50%. Have you met such problems? @joaoluiscarreira
<https://github.com/joaoluiscarreira>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#46 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AO6qaph7hvuSW91qD5DMfoONVt9O-aA0ks5vEquFgaJpZM4aCQxU>
.
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.