首页/论文/2010.02178

arXiv 2010.021782020-10-05cs.CVcs.AIstat.ML

Mind the Pad -- CNNs can Develop Blind Spots

Facebook AI（脸书人工智能研究院）；NYU（纽约大学）

Bilal Alsallakh, Narine Kokhlikyan, Vivek Miglani, Jun Yuan, Orion Reblitz-Richardson

备注：: Appendix E available at https://drive.google.com/file/d/1bIvRQJIBwJbKTfpg0hNaFX2ThuuDO8PU/view?usp=sharing

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

英文摘要相关论文

英文摘要：

We show how feature maps in convolutional networks are susceptible to spatial bias. Due to a combination of architectural choices, the activation at certain locations is systematically elevated or weakened. The major source of this bias is the padding mechanism. Depending on several aspects of convolution arithmetic, this mechanism can apply the padding unevenly, leading to asymmetries in the learned weights. We demonstrate how such bias can be detrimental to certain tasks such as small object detection: the activation is suppressed if the stimulus lies in the impacted area, leading to blind spots and misdetection. We propose solutions to mitigate spatial bias and demonstrate how they can improve model accuracy.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Interpolation-Aware Padding for 3D Sparse Convolutional Neural Networks

Tsinghua University（清华大学）；Microsoft Research Asia（微软亚洲研究院）

Sparse voxel-based 3D convolutional neural networks (CNNs) are widely used for various 3D vision tasks. Sparse voxel-based 3D CNNs create sparse non-empty voxels from the 3D input and perform 3D convolution operations on them only. We propose a simple yet effective padding scheme --- interpolation-aware padding to pad a few empty voxels adjacent to the non-empty voxels and involve them in the 3D CNN computation so that all neighboring voxels exist when computing point-wise features via the trilinear interpolation. For fine-grained 3D vision tasks where point-wise features are essential, like s…

2021-08-16cs.CV

How Can CNNs Use Image Position for Segmentation?

Tohoku University（东北大学）；RIKEN AIP（理化学研究所人工智能研究中心）

Convolution is an equivariant operation, and image position does not affect its result. A recent study shows that the zero-padding employed in convolutional layers of CNNs provides position information to the CNNs. The study further claims that the position information enables accurate inference for several tasks, such as object recognition, segmentation, etc. However, there is a technical issue with the design of the experiments of the study, and thus the correctness of the claim is yet to be verified. Moreover, the absolute image position may not be essential for the segmentation of natural …

2020-05-07cs.CVeess.IV

基于保留核的微分特征的免填充卷积

Padding-free Convolution based on Preservation of Differential Characteristics of Kernels

Science and Technology Facilities Council（科学技术设施委员会）

提出一种基于保留核微分特征的无填充保尺寸卷积方法，通过将不完整窗口卷积坍缩为局部线性微分算子，有效消除边界效应，并在物理场滤波与CNN图像分类等任务中展现明显优势。

2023-09-12cs.CV

Improved Padding in CNNs for Quantitative Susceptibility Mapping

Recently, deep learning methods have been proposed for quantitative susceptibility mapping (QSM) data processing: background field removal, field-to-source inversion, and single-step QSM reconstruction. However, the conventional padding mechanism used in convolutional neural networks (CNNs) can introduce spatial artifacts, especially in QSM background field removal and single-step QSM which requires inference from total fields with extreme large values at the edge boundaries of volume of interest. To address this issue, we propose an improved padding technique which utilizes the neighboring va…

2021-06-21cs.CVeess.IV

Position, Padding and Predictions: A Deeper Look at Position Information in CNNs

Ryerson University（瑞尔森大学）；University of Waterloo（滑铁卢大学）；University of Guelph（圭尔夫大学）；Vector Institute of Artificial Intelligence（向量人工智能研究院）；Samsung AI Centre Toronto（三星多伦多人工智能中心）

In contrast to fully connected networks, Convolutional Neural Networks (CNNs) achieve efficiency by learning weights associated with local filters with a finite spatial extent. An implication of this is that a filter may know what it is looking at, but not where it is positioned in the image. In this paper, we first test this hypothesis and reveal that a surprising degree of absolute position information is encoded in commonly used CNNs. We show that zero padding drives CNNs to encode position information in their internal representations, while a lack of padding precludes position encoding. T…

2021-01-28cs.CV

PadChannel：通过显式填充编码提升CNN性能

PadChannel: Improving CNN Performance through Explicit Padding Encoding

Faculty of Applied Science and Engineering（应用科学与工程学院）；University of Toronto（多伦多大学）

提出PadChannel填充方法，通过额外输入通道编码填充状态，使CNN区分真实与填充像素，在ImageNet-1K上以微小计算成本提升性能并降低方差。

2023-11-16cs.CV

更多相关论文收起更多论文

The Mind's Eye: Visualizing Class-Agnostic Features of CNNs

Utrecht University（乌得勒支大学）

Visual interpretability of Convolutional Neural Networks (CNNs) has gained significant popularity because of the great challenges that CNN complexity imposes to understanding their inner workings. Although many techniques have been proposed to visualize class features of CNNs, most of them do not provide a correspondence between inputs and the extracted features in specific layers. This prevents the discovery of stimuli that each layer responds better to. We propose an approach to visually interpret CNN features given a set of images by creating corresponding images that depict the most inform…

2021-01-29cs.CV

RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation

The University of Adelaide（阿德莱德大学）；Australian Centre for Robotic Vision（澳大利亚机器人视觉中心）

Recently, very deep convolutional neural networks (CNNs) have shown outstanding performance in object recognition and have also been the first choice for dense classification problems such as semantic segmentation. However, repeated subsampling operations like pooling or convolution striding in deep CNNs lead to a significant decrease in the initial image resolution. Here, we present RefineNet, a generic multi-path refinement network that explicitly exploits all the information available along the down-sampling process to enable high-resolution prediction using long-range residual connections.…

2016-11-25cs.CV

用于皮肤癌检测的CNN、Transformer、混合模型和视觉语言模型

CNNs, Transformers, Hybrid, and Vision Language Models for Skin Cancer Detection

Department of Computer Science and Software Engineering, Concordia University, Montreal, Canada（计算机科学与软件工程系，康科迪亚大学，加拿大蒙特利尔）；Ebovir Biotechnologie Inc., Montreal, Canada（Ebovir生物技术公司，加拿大蒙特利尔）

本文在PAD-UFES-20数据集上统一评估了12种深度学习模型（包括CNN、ViT、混合卷积Transformer和视觉语言模型），结果表明混合模型和基于SigLIP的VLM在排名性能和临床相关操作点之间取得了最佳平衡。

2026-05-25cs.CV

Ensemble of Multi-View Learning Classifiers for Cross-Domain Iris Presentation Attack Detection

University of Notre Dame（圣母大学）；University of Campinas (Unicamp)（坎皮纳斯大学（Unicamp））

The adoption of large-scale iris recognition systems around the world has brought to light the importance of detecting presentation attack images (textured contact lenses and printouts). This work presents a new approach in iris Presentation Attack Detection (PAD), by exploring combinations of Convolutional Neural Networks (CNNs) and transformed input spaces through binarized statistical image features (BSIF). Our method combines lightweight CNNs to classify multiple BSIF views of the input image. Following explorations on complementary input spaces leading to more discriminative features to d…

2018-11-25cs.CV

Analysis of Real-Time Hostile Activitiy Detection from Spatiotemporal Features Using Time Distributed Deep CNNs, RNNs and Attention-Based Mechanisms

BRAC University（BRAC大学）；University of Delaware（特拉华大学）

Real-time video surveillance, through CCTV camera systems has become essential for ensuring public safety which is a priority today. Although CCTV cameras help a lot in increasing security, these systems require constant human interaction and monitoring. To eradicate this issue, intelligent surveillance systems can be built using deep learning video classification techniques that can help us automate surveillance systems to detect violence as it happens. In this research, we explore deep learning video classification techniques to detect violence as they are happening. Traditional image classi…

2023-02-21cs.CV