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.

Attention:

The NSF Public Access Repository (PAR) system and access will be unavailable from 11:00 PM ET on Friday, May 2 until 12:00 AM ET on Saturday, May 3 due to maintenance. We apologize for the inconvenience.

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


Title: Position-Aware Recalibration Module: Learning From Feature Semantics and Feature Position
We present a new method to improve the representational power of the features in Convolutional Neural Networks (CNNs). By studying traditional image processing methods and recent CNN architectures, we propose to use positional information in CNNs for effective exploration of feature dependencies. Rather than considering feature semantics alone, we incorporate spatial positions as an augmentation for feature semantics in our design. From this vantage, we present a Position-Aware Recalibration Module (PRM in short) which recalibrates features leveraging both feature semantics and position. Furthermore, inspired by multi-head attention, our module is capable of performing multiple recalibrations where results are concatenated as the output. As PRM is efficient and easy to implement, it can be seamlessly integrated into various base networks and applied to many position-aware visual tasks. Compared to original CNNs, our PRM introduces a negligible number of parameters and FLOPs, while yielding better performance. Experimental results on ImageNet and MS COCO benchmarks show that our approach surpasses related methods by a clear margin with less computational overhead. For example, we improve the ResNet50 by absolute 1.75% (77.65% vs. 75.90%) on ImageNet 2012 validation dataset, and 1.5%~1.9% mAP on MS COCO validation dataset with almost no computational overhead. Codes are made publicly available.  more » « less
Award ID(s):
1852134
PAR ID:
10206340
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence
Page Range / eLocation ID:
797 to 803
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, 16th European Conference on Computer Vision (2020))
    Vedaldi, A. (Ed.)
    In state-of-the-art deep neural networks, both feature normalization and feature attention have become ubiquitous. They are usually studied as separate modules, however. In this paper, we propose a light-weight integration between the two schema and present Attentive Normalization (AN). Instead of learning a single affine transformation, AN learns a mixture of affine transformations and utilizes their weighted-sum as the final affine transformation applied to re-calibrate features in an instance-specific way. The weights are learned by leveraging channel-wise feature attention. In experiments, we test the proposed AN using four representative neural architectures. In the ImageNet-1000 classification benchmark and the MS-COCO 2017 object detection and instance segmentation benchmark. AN obtains consistent performance improvement for different neural architectures in both benchmarks with absolute increase of top-1 accuracy in ImageNet-1000 between 0.5\% and 2.7\%, and absolute increase up to 1.8\% and 2.2\% for bounding box and mask AP in MS-COCO respectively. We observe that the proposed AN provides a strong alternative to the widely used Squeeze-and-Excitation (SE) module. The source codes are publicly available at \href{https://github.com/iVMCL/AOGNet-v2}{the ImageNet Classification Repo} and \href{https://github.com/iVMCL/AttentiveNorm\_Detection}{the MS-COCO Detection and Segmentation Repo}. 
    more » « less
  2. ; ; ; ; ; (, IEEE)
    Vision Transformers (ViTs) are built on the assumption of treating image patches as “visual tokens” and learn patch-to-patch attention. The patch embedding based tokenizer has a semantic gap with respect to its counterpart, the textual tokenizer. The patch-to-patch attention suffers from the quadratic complexity issue, and also makes it non-trivial to explain learned ViTs. To address these issues in ViT, this paper proposes to learn Patch-to-Cluster attention (PaCa) in ViT. Queries in our PaCa-ViT starts with patches, while keys and values are directly based on clustering (with a predefined small number of clusters). The clusters are learned end-to-end, leading to better tokenizers and inducing joint clustering-for-attention and attention-for-clustering for better and interpretable models. The quadratic complexity is relaxed to linear complexity. The proposed PaCa module is used in designing efficient and interpretable ViT backbones and semantic segmentation head networks. In experiments, the proposed methods are tested on ImageNet-1k image classification, MS-COCO object detection and instance segmentation and MIT-ADE20k semantic segmentation. Compared with the prior art, it obtains better performance in all the three benchmarks than the SWin [32] and the PVTs [47], [48] by significant margins in ImageNet-1k and MIT-ADE20k. It is also significantly more efficient than PVT models in MS-COCO and MIT-ADE20k due to the linear complexity. The learned clusters are semantically meaningful. Code and model checkpoints are available at https:/github.com/iVMCL/PaCaViT. 
    more » « less
  3. ; ; ; ; (, 2021 Third International Conference on Transdisciplinary AI (TransAI))
    An efficient feature selection method can significantly boost results in classification problems. Despite ongoing improvement, hand-designed methods often fail to extract features capturing high- and mid-level representations at effective levels. In machine learning (Deep Learning), recent developments have improved upon these hand-designed methods by utilizing automatic extraction of features. Specifically, Convolutional Neural Networks (CNNs) are a highly successful technique for image classification which can automatically extract features, with ongoing learning and classification of these features. The purpose of this study is to detect hydraulic structures (i.e., bridges and culverts) that are important to overland flow modeling and environmental applications. The dataset used in this work is a relatively small dataset derived from 1-m LiDAR-derived Digital Elevation Models (DEMs) and National Agriculture Imagery Program (NAIP) aerial imagery. The classes for our experiment consist of two groups: the ones with a bridge/culvert being present are considered "True", and those without a bridge/culvert are considered "False". In this paper, we use advanced CNN techniques, including Siamese Neural Networks (SNNs), Capsule Networks (CapsNets), and Graph Convolutional Networks (GCNs), to classify samples with similar topographic and spectral characteristics, an objective which is challenging utilizing traditional machine learning techniques, such as Support Vector Machine (SVM), Gaussian Classifier (GC), and Gaussian Mixture Model (GMM). The advanced CNN-based approaches combined with data pre-processing techniques (e.g., data augmenting) produced superior results. These approaches provide efficient, cost-effective, and innovative solutions to the identification of hydraulic structures. 
    more » « less
  4. ; ; ; ; (, International Conference on Pattern Recognition (ICPR))
    In this paper, we propose a deep multimodal fusion network to fuse multiple modalities (face, iris, and fingerprint) for person identification. The proposed deep multimodal fusion algorithm consists of multiple streams of modality-specific Convolutional Neural Networks (CNNs), which are jointly optimized at multiple feature abstraction levels. Multiple features are extracted at several different convolutional layers from each modality-specific CNN for joint feature fusion, optimization, and classification. Features extracted at different convolutional layers of a modality-specific CNN represent the input at several different levels of abstract representations. We demonstrate that an efficient multimodal classification can be accomplished with a significant reduction in the number of network parameters by exploiting these multi-level abstract representations extracted from all the modality-specific CNNs. We demonstrate an increase in multimodal person identification performance by utilizing the proposed multi-level feature abstract representations in our multimodal fusion, rather than using only the features from the last layer of each modality-specific CNNs. We show that our deep multi-modal CNNs with multimodal fusion at several different feature level abstraction can significantly outperform the unimodal representation accuracy. We also demonstrate that the joint optimization of all the modality-specific CNNs excels the score and decision level fusions of independently optimized CNNs. 
    more » « less
  5. ; ; (, The 17th IEEE International Conference on Machine Learning and Applications (ICMLA))
    Convolutional neural networks (CNN) have become very popular for computer vision, text, and sequence tasks. CNNs have the advantage of being able to learn local patterns through convolution filters. However, generic datasets do not have meaningful local data correlations, because their features are assumed to be independent of each other. In this paper, we propose an approach to reorder features of a generic dataset to create feature correlations for CNN to learn feature representation, and use learned features as inputs to help improve traditional machine learning classifiers. Our experiments on benchmark data exhibit increased performance and illustrate the benefits of using CNNs for generic datasets. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email