skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Temporal Gaussian Mixture Layer for Videos
We introduce a new convolutional layer named the Temporal Gaussian Mixture (TGM) layer and present how it can be used to efficiently capture longer-term temporal information in continuous activity videos. The TGM layer is a temporal convolutional layer governed by a much smaller set of parameters (e.g., location/variance of Gaussians) that are fully differentiable. We present our fully convolutional video models with multiple TGM layers for activity detection. The extensive experiments on multiple datasets, including Charades and MultiTHUMOS, confirm the effectiveness of TGM layers, significantly outperforming the state-of-the-arts.  more » « less
Award ID(s):
1812943 1814985
PAR ID:
10107544
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of Machine Learning Research
Volume:
97
ISSN:
2640-3498
Page Range / eLocation ID:
5152-5161
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; (, Computer Vision and Image Understanding)
    Weakly Supervised Semantic Segmentation (WSSS) provides efficient solutions for semantic image segmentation using image-level annotations. WSSS requires no pixel-level labeling that Fully Supervised Semantic Segmentation (FSSS) does, which is time-consuming and label-intensive. Most WSSS approaches have leveraged Class Activation Maps (CAM) or Self-Attention (SA) to generate pseudo pixel-level annotations to perform semantic segmentation tasks coupled with fully supervised approaches (e.g., Fully Convolutional Network). However, those approaches often provides incomplete supervision that mainly includes discriminative regions from the last convolutional layer. They may fail to capture regions of low- or intermediate-level features that may not be present in the last convolutional layer. To address the issue, we proposed a novel Multi-layered Self-Attention (Multi-SA) method that applies a self-attention module to multiple convolutional layers, and then stack feature maps from the self-attention layers to generate pseudo pixel-level annotations. We demonstrated that integrated feature maps from multiple self-attention layers produce higher coverage in semantic segmentation than using only the last convolutional layer through intensive experiments using standard benchmark datasets. 
    more » « less
  2. ; (, Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition)
    In this paper, we propose a convolutional layer inspired by optical flow algorithms to learn motion representations. Our representation flow layer is a fully-differentiable layer designed to capture the `flow' of any representation channel within a convolutional neural network for action recognition. Its parameters for iterative flow optimization are learned in an end-to-end fashion together with the other CNN model parameters, maximizing the action recognition performance. Furthermore, we newly introduce the concept of learning `flow of flow' representations by stacking multiple representation flow layers. We conducted extensive experimental evaluations, confirming its advantages over previous recognition models using traditional optical flows in both computational speed and performance. The code is publicly available. 
    more » « less
  3. ; (, PLOS computational biology)
    Visual scene category representations emerge very rapidly, yet the computational transformations that enable such invariant categorizations remain elusive. Deep convolutional neural networks (CNNs) perform visual categorization at near human-level accuracy using a feedforward architecture, providing neuroscientists with the opportunity to assess one successful series of representational transformations that enable categorization in silico. The goal of the current study is to assess the extent to which sequential scene category representations built by a CNN map onto those built in the human brain as assessed by high-density, time-resolved event-related potentials (ERPs). We found correspondence both over time and across the scalp: earlier (0–200 ms) ERP activity was best explained by early CNN layers at all electrodes. Although later activity at most electrode sites corresponded to earlier CNN layers, activity in right occipito-temporal electrodes was best explained by the later, fully-connected layers of the CNN around 225 ms post-stimulus, along with similar patterns in frontal electrodes. Taken together, these results suggest that the emergence of scene category representations develop through a dynamic interplay between early activity over occipital electrodes as well as later activity over temporal and frontal electrodes. 
    more » « less
  4. ; (, 6th World Congress on Global Optimization)
    We study sparsification of convolutional neural networks (CNN) by a relaxed variable splitting method of ℓ0 and transformed-ℓ1 (Tℓ1) penalties, with application to complex curves such as texts written in different fonts, and words written with trembling hands simulating those of Parkinson’s disease patients. The CNN contains 3 convolutional layers, each followed by a maximum pooling, and finally a fully connected layer which contains the largest number of network weights. With ℓ0 penalty, we achieved over 99% test accuracy in distinguishing shaky vs. regular fonts or hand writings with above 86% of the weights in the fully connected layer being zero. Comparable sparsity and test accuracy are also reached with a proper choice of Tℓ1 penalty. 
    more » « less
  5. (, 2018 IEEE Conf. on Computer Vision and Pattern Recognition.)
    Bilinear pooling has been recently proposed as a feature encoding layer, which can be used after the convolutional layers of a deep network, to improve performance in multiple vision tasks. Different from conventional global average pooling or fully connected layer, bilinear pooling gathers 2nd order information in a translation invariant fashion. However, a serious drawback of this family of pooling layers is their dimensionality explosion. Approximate pooling methods with compact properties have been explored towards resolving this weakness. Additionally, recent results have shown that significant performance gains can be achieved by adding 1st order information and applying matrix normalization to regularize unstable higher order information. However, combining compact pooling with matrix normalization and other order information has not been explored until now. In this paper, we unify bilinear pooling and the global Gaussian embedding layers through the empirical moment matrix. In addition, we propose a novel sub-matrix square-root layer, which can be used to normalize the output of the convolution layer directly and mitigate the dimensionality problem with off-the-shelf compact pooling methods. Our experiments on three widely used finegrained classification datasets illustrate that our proposed architecture, MoNet, can achieve similar or better performance than with the state-of-art G2DeNet. Furthermore, when combined with compact pooling technique, MoNet obtains comparable performance with encoded features with 96% less dimensions. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email