Search Results for author: Sandra Avila

Found 29 papers, 22 papers with code

Even Small Correlation and Diversity Shifts Pose Dataset-Bias Issues

no code implementations9 May 2023 Alceu Bissoto, Catarina Barata, Eduardo Valle, Sandra Avila

Our protocol reveals three findings: 1) Models learn and propagate correlation shifts even with low-bias training; this poses a risk of accumulating and combining unaccountable weak biases; 2) Models learn robust features in high- and low-bias scenarios but use spurious ones if test samples have them; this suggests that spurious correlations do not impair the learning of robust features; 3) Diversity shift can reduce the reliance on spurious correlations; this is counter intuitive since we expect biased models to depend more on biases when invariant features are missing.

Artifact-Based Domain Generalization of Skin Lesion Models

1 code implementation20 Aug 2022 Alceu Bissoto, Catarina Barata, Eduardo Valle, Sandra Avila

We propose a pipeline that relies on artifacts annotation to enable generalization evaluation and debiasing for the challenging skin lesion analysis context.

Domain Generalization Out-of-Distribution Generalization

A Survey on Deep Learning for Skin Lesion Segmentation

1 code implementation1 Jun 2022 Zahra Mirikharaji, Kumar Abhishek, Alceu Bissoto, Catarina Barata, Sandra Avila, Eduardo Valle, M. Emre Celebi, Ghassan Hamarneh

We analyze these works along several dimensions, including input data (datasets, preprocessing, and synthetic data generation), model design (architecture, modules, and losses), and evaluation aspects (data annotation requirements and segmentation performance).

Lesion Segmentation Skin Lesion Segmentation +1

Seeing without Looking: Analysis Pipeline for Child Sexual Abuse Datasets

no code implementations29 Apr 2022 Camila Laranjeira, João Macedo, Sandra Avila, Jefersson A. dos Santos

The online sharing and viewing of Child Sexual Abuse Material (CSAM) are growing fast, such that human experts can no longer handle the manual inspection.

Pornography Detection

Weakly Supervised Attention-based Models Using Activation Maps for Citrus Mite and Insect Pest Classification

no code implementations2 Oct 2021 Edson Bollis, Helena Maia, Helio Pedrini, Sandra Avila

Despite the integrated pest mechanical aspect, only a few works on automatic classification have handled images with orange mite characteristics, which means tiny and noisy regions of interest.

Classification Multiple Instance Learning +1

CIDEr-R: Robust Consensus-based Image Description Evaluation

1 code implementation WNUT (ACL) 2021 Gabriel Oliveira dos Santos, Esther Luna Colombini, Sandra Avila

This paper shows that CIDEr-D, a traditional evaluation metric for image description, does not work properly on datasets where the number of words in the sentence is significantly greater than those in the MS COCO Captions dataset.

An Evaluation of Self-Supervised Pre-Training for Skin-Lesion Analysis

1 code implementation17 Jun 2021 Levy Chaves, Alceu Bissoto, Eduardo Valle, Sandra Avila

Self-supervised pre-training appears as an advantageous alternative to supervised pre-trained for transfer learning.

Transfer Learning

GAN-Based Data Augmentation and Anonymization for Skin-Lesion Analysis: A Critical Review

1 code implementation20 Apr 2021 Alceu Bissoto, Eduardo Valle, Sandra Avila

Despite the growing availability of high-quality public datasets, the lack of training samples is still one of the main challenges of deep-learning for skin lesion analysis.

Data Augmentation

#PraCegoVer: A Large Dataset for Image Captioning in Portuguese

2 code implementations21 Mar 2021 Gabriel Oliveira dos Santos, Esther Luna Colombini, Sandra Avila

Thus, inspired by this movement, we have proposed the #PraCegoVer, a multi-modal dataset with Portuguese captions based on posts from Instagram.

Image Captioning TAG

Less is More: Sample Selection and Label Conditioning Improve Skin Lesion Segmentation

1 code implementation28 Apr 2020 Vinicius Ribeiro, Sandra Avila, Eduardo Valle

Segmenting skin lesions images is relevant both for itself and for assisting in lesion classification, but suffers from the challenge in obtaining annotated data.

Experimental Design Lesion Classification +2

Debiasing Skin Lesion Datasets and Models? Not So Fast

1 code implementation23 Apr 2020 Alceu Bissoto, Eduardo Valle, Sandra Avila

Data-driven models are now deployed in a plethora of real-world applications - including automated diagnosis - but models learned from data risk learning biases from that same data.

Lesion Classification Skin Lesion Classification

Weakly Supervised Learning Guided by Activation Mapping Applied to a Novel Citrus Pest Benchmark

2 code implementations22 Apr 2020 Edson Bollis, Helio Pedrini, Sandra Avila

Pests and diseases are relevant factors for production losses in agriculture and, therefore, promote a huge investment in the prevention and detection of its causative agents.

Management Weakly-supervised Learning

The Six Fronts of the Generative Adversarial Networks

no code implementations29 Oct 2019 Alceu Bissoto, Eduardo Valle, Sandra Avila

Generative Adversarial Networks fostered a newfound interest in generative models, resulting in a swelling wave of new works that new-coming researchers may find formidable to surf.

Grape detection, segmentation and tracking using deep neural networks and three-dimensional association

1 code implementation26 Jul 2019 Thiago T. Santos, Leonardo L. de Souza, Andreza A. dos Santos, Sandra Avila

In a test set containing 408 grape clusters from images taken on a trellis-system based vineyard, we have reached an F 1 -score up to 0. 91 for instance segmentation, a fine separation of each cluster from other structures in the image that allows a more accurate assessment of fruit size and shape.

Instance Segmentation Semantic Segmentation

Handling Inter-Annotator Agreement for Automated Skin Lesion Segmentation

1 code implementation6 Jun 2019 Vinicius Ribeiro, Sandra Avila, Eduardo Valle

We also evaluate how conditioning the ground truths using different (but very simple) algorithms may help to enhance agreement and may be appropriate for some use cases.

Image Segmentation Lesion Classification +3

Solo or Ensemble? Choosing a CNN Architecture for Melanoma Classification

1 code implementation29 Apr 2019 Fábio Perez, Sandra Avila, Eduardo Valle

We evaluate that claim for melanoma classification, over 9 CNNs architectures, in 5 sets of splits created on the ISIC Challenge 2017 dataset, and 3 repeated measures, resulting in 135 models.

Classification General Classification +3

Combating the Elsagate phenomenon: Deep learning architectures for disturbing cartoons

1 code implementation18 Apr 2019 Akari Ishikawa, Edson Bollis, Sandra Avila

As the first to explore disturbing content in cartoons, we proceed from the most recent pornography detection literature applying deep convolutional neural networks combined with static and motion information of the video.

Pornography Detection

(De)Constructing Bias on Skin Lesion Datasets

1 code implementation18 Apr 2019 Alceu Bissoto, Michel Fornaciali, Eduardo Valle, Sandra Avila

We fed models with additional clinically meaningful information, which failed to improve the results even slightly, suggesting the destruction of cogent correlations.

BIG-bench Machine Learning

Data Augmentation for Skin Lesion Analysis

1 code implementation5 Sep 2018 Fábio Perez, Cristina Vasconcelos, Sandra Avila, Eduardo Valle

In this work, we investigate the impact of 13 data augmentation scenarios for melanoma classification trained on three CNNs (Inception-v4, ResNet, and DenseNet).

Data Augmentation General Classification +2

Deep-Learning Ensembles for Skin-Lesion Segmentation, Analysis, Classification: RECOD Titans at ISIC Challenge 2018

no code implementations25 Aug 2018 Alceu Bissoto, Fábio Perez, Vinícius Ribeiro, Michel Fornaciali, Sandra Avila, Eduardo Valle

This extended abstract describes the participation of RECOD Titans in parts 1 to 3 of the ISIC Challenge 2018 "Skin Lesion Analysis Towards Melanoma Detection" (MICCAI 2018).

Classification General Classification +2

Data, Depth, and Design: Learning Reliable Models for Skin Lesion Analysis

1 code implementation1 Nov 2017 Eduardo Valle, Michel Fornaciali, Afonso Menegola, Julia Tavares, Flávia Vasques Bittencourt, Lin Tzy Li, Sandra Avila

We use the exhaustive trials to simulate sequential decisions and ensembles, with and without the use of privileged information from the test set.

Data Augmentation Transfer Learning

Knowledge Transfer for Melanoma Screening with Deep Learning

2 code implementations22 Mar 2017 Afonso Menegola, Michel Fornaciali, Ramon Pires, Flávia Vasques Bittencourt, Sandra Avila, Eduardo Valle

Knowledge transfer impacts the performance of deep learning -- the state of the art for image classification tasks, including automated melanoma screening.

Image Classification Skin Cancer Classification +1

RECOD Titans at ISIC Challenge 2017

4 code implementations14 Mar 2017 Afonso Menegola, Julia Tavares, Michel Fornaciali, Lin Tzy Li, Sandra Avila, Eduardo Valle

This extended abstract describes the participation of RECOD Titans in parts 1 and 3 of the ISIC Challenge 2017 "Skin Lesion Analysis Towards Melanoma Detection" (ISBI 2017).

General Classification Lesion Segmentation +2

A Mid-level Video Representation based on Binary Descriptors: A Case Study for Pornography Detection

1 code implementation12 May 2016 Carlos Caetano, Sandra Avila, William Robson Schwartz, Silvio Jamil F. Guimarães, Arnaldo de A. Araújo

Although these approaches provide good results, they generally have the disadvantage of a high false positive rate since not all images with large areas of skin exposure are necessarily pornographic images, such as people wearing swimsuits or images related to sports.

Pornography Detection Video Description

Deep Neural Networks Under Stress

1 code implementation11 May 2016 Micael Carvalho, Matthieu Cord, Sandra Avila, Nicolas Thome, Eduardo Valle

In recent years, deep architectures have been used for transfer learning with state-of-the-art performance in many datasets.

Transfer Learning

Towards Automated Melanoma Screening: Proper Computer Vision & Reliable Results

2 code implementations14 Apr 2016 Michel Fornaciali, Micael Carvalho, Flávia Vasques Bittencourt, Sandra Avila, Eduardo Valle

In this paper we survey, analyze and criticize current art on automated melanoma screening, reimplementing a baseline technique, and proposing two novel ones.

Melanoma Diagnosis

Semantic Diversity versus Visual Diversity in Visual Dictionaries

no code implementations20 Nov 2015 Otávio A. B. Penatti, Sandra Avila, Eduardo Valle, Ricardo da S. Torres

Results for image classification show that as visual dictionaries are based on low-level visual appearances, visual diversity is more important than semantic diversity.

General Classification Image Classification +1

Cannot find the paper you are looking for? You can Submit a new open access paper.