paper-with-me

Papers

Eigen-Distortions of Hierarchical Representations

2017-10-06 · NeurIPS 2017 12 · Alexander Berardino, Johannes Ballé, Valero Laparra, Eero P. Simoncelli

We develop a method for comparing hierarchical image representations in terms of their ability to explain perceptual sensitivity in humans. Specifically, we utilize Fisher information to establish a model-derived prediction of sensitivity to local perturbations of an image. For a given image, we compute the eigenvectors of the Fisher information matrix with largest and smallest eigenvalues, corresponding to the model-predicted most- and least-noticeable image distortions, respectively. For human subjects, we then measure the amount of each distortion that can be reliably detected when added to the image. We use this method to test the ability of a variety of representations to mimic human perceptual sensitivity. We find that the early layers of VGG16, a deep neural network optimized for object recognition, provide a better match to human perception than later layers, and a better match than a 4-stage convolutional neural network (CNN) trained on a database of human ratings of distorted image quality. On the other hand, we find that simple models of early visual processing, incorporating one or more stages of local gain control, trained on the same database of distortion ratings, provide substantially better predictions of human sensitivity than either the CNN, or any combination of layers of VGG16.

📄 PDF Abstract BibTeX arXiv:1710.02266

Code (0)

등록된 구현이 없습니다.

Tasks

Object RecognitionSensitivity

Similar Papers 제목 키워드 기반

A Geometric Perspective for High-Dimensional Multiplex Graphs

2025-01-29 · Kamel Abdous, Nairouz Mrabah, Mohamed Bouguessa

High-dimensional multiplex graphs are characterized by their high number of complementary and divergent dimensions. The existence of multiple hierarchical latent relations between the graph dimensions poses significant c…

Graph Embedding

Eigenoption Discovery through the Deep Successor Representation

2017-10-30 · ICLR 2018 1 · Marlos C. Machado, Clemens Rosenbaum, Xiaoxiao Guo, Miao Liu 외

Options in reinforcement learning allow agents to hierarchically decompose a task into subtasks, having the potential to speed up learning and planning. However, autonomously learning effective sets of options is still a…

Atari GamesDeep Reinforcement Learningreinforcement-learningReinforcement Learning+2

Hyperbolic Heterogeneous Graph Attention Networks

2024-04-15 · Jongmin Park, SeungHoon Han, Soohwan Jeong, Sungsu Lim

Most previous heterogeneous graph embedding models represent elements in a heterogeneous graph as vector representations in a low-dimensional Euclidean space. However, because heterogeneous graphs inherently possess comp…

ClusteringGraph AttentionGraph EmbeddingNode Classification

Disentangling Image Distortions in Deep Feature Space

2020-02-26 · Simone Bianco, Luigi Celona, Paolo Napoletano

Previous literature suggests that perceptual similarity is an emergent property shared across deep visual representations. Experiments conducted on a dataset of human-judged image distortions have proven that deep featur…

Image Quality Assessment

On the Spectral Bias of Convolutional Neural Tangent and Gaussian Process Kernels

2022-03-17 · Amnon Geifman, Meirav Galun, David Jacobs, Ronen Basri

We study the properties of various over-parametrized convolutional neural architectures through their respective Gaussian process and neural tangent kernels. We prove that, with normalized multi-channel input and ReLU ac…