Will Transfer Learning Enhance ImageNet Classification Accuracy Using ImageNet-Pretrained Models?

The huge impact of Transfer Learning (TL) techniques in many fields was achieved using several state-of-the-art ImageNet-pretrained models. These models have shown great performance improvements on this dataset over the last few years. One of the recently used TL techniques is feature extraction with the help of Feature Concatenation (FC), where the extracted features of multiple pretrained models are concatenated together before training on them, to produce more robust and discriminative feature representations on various classification tasks. However, neither TL nor FC techniques have been tested on the same dataset that initially trained the pretrained models, i.e. ImageNet. Hence, this work provides an investigative study to test the possibility of improving the ImageNet accuracy using the feature extraction approach of TL with the help of FC techniques. The results of this work show that there is no TL technique that can be used with or without FC to increase the accuracy of pretrained models on the original dataset on which they were trained. Even for FC, it cannot produce a more discriminative feature representation for the original data than what the individual models can produce.

• Publication year

  2019

• Venue

  International Conference on Information, Communications and Signal Processing

• Publication date

  2019-06-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1109/IACS.2019.8809114
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Performance study of augmentation techniques for HEp2 CNN classification

  2018cited by this paper
• Deep CNNs for microscopic image classification by exploiting transfer learning and feature concatenation

  2018cited by this paper
• Convolutional Neural Network Ensemble Fine-Tuning for Extended Transfer Learning

  2018cited by this paper
• Do Better ImageNet Models Transfer Better?

  2018influential reference
• Determine Bipolar Disorder Level from Patient Interviews Using Bi-LSTM and Feature Fusion

  2018cited by this paper
• Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

  2016cited by this paper
• Identity Mappings in Deep Residual Networks

  2016cited by this paper
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size

  2016cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Rethinking the Inception Architecture for Computer Vision

  2015cited by this paper
• Going deeper with convolutions

  2014cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• ImageNet Large Scale Visual Recognition Challenge

  2014influential reference
• CNN Features Off-the-Shelf: An Astounding Baseline for Recognition

  2014cited by this paper
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition

  2013cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• ImageNet: A large-scale hierarchical image database

  2009cited by this paper
• A dual coordinate descent method for large-scale linear SVM

  2008cited by this paper
• Links between perceptrons, MLPs and SVMs

  2004cited by this paper
• Support-Vector Networks

  1995cited by this paper
• Parallel Networks that Learn to Pronounce English Text

  1987cited by this paper
• Learning Translation Invariant Recognition in Massively Parallel Networks

  1987cited by this paper
• Learning internal representations by error propagation

  1986cited by this paper
• PRINCIPLES OF NEURODYNAMICS. PERCEPTRONS AND THE THEORY OF BRAIN MECHANISMS

  1963cited by this paper
• Receptive fields of single neurones in the cat's striate cortex

  1959influential reference

• Effective SMOTE boost with deep learning for IDC identification in whole-slide images

  2025cites this paper
• Enhancing fog load balancing through lifelong transfer learning of reinforcement learning agents

  2024cites this paper
• Supervised Deep Learning for Ideal Identification of Image Retargeting Techniques

  2024cites this paper
• Advanced disk herniation computer aided diagnosis system

  2024cites this paper
• Corrosion Prediction in the Oil Industry Using Deep Learning Techniques

  2024cites this paper
• Breast histopathological imaging using ultra‐fast fluorescence confocal microscopy to identify cancer lesions at early stage

  2024cites this paper
• Facial expression recognition (FER) survey: a vision, architectural elements, and future directions

  2024cites this paper
• Fully Distributed Fog Load Balancing with Multi-Agent Reinforcement Learning

  2024cites this paper
• Quality Feature Learning via Multi-Channel CNN and GRU for No-Reference Video Quality Assessment

  2023cites this paper
• Image Retargeting Techniques Identification Using Supervised Deep Learning

  2023cites this paper
• Fingerprint Identification from Digital Images Using Deep Learning

  2023cites this paper
• Lifelong Learning for Fog Load Balancing: A Transfer Learning Approach

  2023cites this paper
• Development and integration of VGG and dense transfer-learning systems supported with diverse lung images for discovery of the Coronavirus identity Informatics in Medicine Unlocked

  2022cites this paper
• Prostate cancer classification from ultrasound and MRI images using deep learning based Explainable Artificial Intelligence

  2022cites this paper
• Development and integration of VGG and dense transfer-learning systems supported with diverse lung images for discovery of the Coronavirus identity

  2022influential citation
• Automatic classification of images with beach linear perspective using convolutional neural networks

  2022cites this paper
• Transfer learning en la clasificación binaria de imágenes térmicas

  2021cites this paper
• Transfer Learning for Binary Classification of Thermal Images Transfer learning en la clasificación binaria de imágenes térmicas

  2021cites this paper
• Blind 360-degree image quality assessment via saliency-guided convolution neural network

  2021cites this paper
• Building Damage Detection Using U-Net with Attention Mechanism from Pre- and Post-Disaster Remote Sensing Datasets

  2021cites this paper
• An Efficient Face Mask Detector with PyTorch and Deep Learning

  2021influential citation
• 2ST-UNet: 2-Stage Training Model using U-Net for Pneumothorax Segmentation in Chest X-Rays

  2020cites this paper
• The 2ST-UNet for Pneumothorax Segmentation in Chest X-Rays using ResNet34 as a Backbone for U-Net

  2020cites this paper
• An automatic sorting system for unwashed eggs using deep learning

  2020cites this paper
• An Efficienti Fabric Pattern Classification Using Transfer Learning Convolutional Network

  year unknowncites this paper