Comments (3)
Thanks @LiliMeng for the feedback and trying out attentional pooling on CIFAR-10!
Here's what I think might be happening here:
- Given the connection (low-rank approx) we show to 2nd order pooling, a variant of which has been shown to be useful for fine-grained recognition (bilinear cnn), and concurrently for other fine-grained tasks like VQA (relation nets), the 10-way classification task on CIFAR might not be "fine-grained" enough.
- Again using the connection to 2nd order pooling, I think attentional pooling would be most useful for tasks that require "interaction" features ($X^TX$); i.e. pair-wise features of one part of the image/video with another. This I think is especially true in action recognition and VQA, which might explain the recent success of similar self-attention methods for these tasks (Attend and Interact, Non-local Neural Networks)
- The resolution of images is also important. If your base network has high receptive field, you might want to increase the resolution of the input images (simple resizing would work), so that the last layer neurons look at different regions of the image, and attentional pooling can down-weight certain regions.
That said, our attention module is super light-weight, and seems to mostly maintain or improve performance; it might be useful to keep around in your network architectures
from attentionalpoolingaction.
Thanks a lot @rohitgirdhar for your kind and detailed reply! :)
-
I also tried on attentional pooling for CIFAR100 with ResNet-32, the result is: 69.44% (with attentional pooling) vs.70.05% (without). Although CIAFR100 has 10 super classes, each super class has 10 sub-classes, such as rabbit and squirrel. Maybe it is not "fine-grained" as well? The CIFAR10/100 feature map before the pooling layer is [batch_size, 8, 8, 64]. Maybe 8x8 is too small for weighting certain regions. Or because the CIFAR10/100, the object needed to recognize have already taken up most space of the image, additional attention may not be helpful?
-
Is resolution of images also important? I also tried on STL10, it's 96x96 (three times of CIFAR10/100), the result is 79.69% (with attentional pooling) Vs. 80.92% (without).
-
I'll have a try to use large images and activity datasets.
from attentionalpoolingaction.
Thanks for trying the other experiments. Yes, the resolution of the image is quite important, as you want different features at the last layer focus on different areas of the image.
from attentionalpoolingaction.
Related Issues (20)
- Error message -- undefined symbol: _ZN2cv8fastFreeEPv HOT 2
- Error: /usr/bin/ld: cannot find -lboost_system HOT 3
- Infer realtime HOT 1
- HICO and HMDB51 dataset HOT 3
- Results of self training + eval are lower than reported HOT 2
- Error while compiling the custom ops HOT 1
- Parameters/HyperParameters for training HICO dataset HOT 2
- File not found: ../experiments/002_MPII_ResNet_withAttention_train+val.yaml HOT 1
- Some implementation queries on dropout, vgg16 last feature map tapping HOT 1
- Training issue about another dataset
- How to install the IlmImf and extra_component library HOT 1
- undefined symbol: _ZN5boost13thread_detail20rollback_once_regionERNS_9once_flagE HOT 3
- Cannot find libs folder for tensorflow_compact_bilinear_pooling HOT 2
- ValueError: need at least one array to concatenate HOT 1
- Hello, can you write an Attentional Pooling version that caffe can use? HOT 1
- Error when running any one of the python files HOT 1
- Failed to find any matching file for resnet_v1_101.ckpt HOT 3
- undefined symbol: _ZTIN10tensorflow8OpKernelE HOT 4
- Classification Output Issue
- attentional pooling layer 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 attentionalpoolingaction.