Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks100 from public data
- Image annotation using SVM314
An innovative image annotation tool for classifying image regions in one of seven classes - sky, skin, vegetation, snow, water, ground, and buildings - or as unknown is described.
- 3D face detection using curvature analysis255
This work presents an innovative method that combines a feature-based approach with a holistic one for three-dimensional (3D) face detection, which has been tested, with good results, on some 150 3D faces acquired by a laser range scanner.
- Color constancy using CNNs232
Taking image patches as input, the CNN works in the spatial domain without using hand-crafted features that are employed by most previous methods to accurately predict the scene illumination.
- Single and Multiple Illuminant Estimation Using Convolutional Neural Networks168
An extensive comparison with both local and global illuminant estimation methods in the state of the art, on standard data sets with single and multiple illuminants, proves the effectiveness of this three-stage method.
- 150
- Automatic color constancy algorithm selection and combination119
This work investigates how illuminant estimation techniques can be improved taking into account intrinsic, low level properties of the images, and shows how these properties can be used to drive the selection of the best algorithm for a given image.
- 111
- 102
- 98
- 68
- Evaluating color texture descriptors under large variations of controlled lighting conditions67
An extensive comparison of old and new texture features, with and without a color normalization step, is reported, with a particular focus on how these features are affected by small and large variations in the lighting conditions.
- 49
- Artistic photo filter removal using convolutional neural networks48
The method uses a convolutional neural network for the prediction of the coefficients of local polynomial transformations that are applied to the input image and shows that the quality of the restoration performed by the method is clearly superior to that of traditional color balancing and restoration procedures.
- Three-Dimensional Occlusion Detection and Restoration of Partially Occluded Faces48
This paper presents an innovative three dimensional occlusion detection and restoration strategy for the recognition of three dimensional faces partially occluded by unforeseen, extraneous objects and demonstrates the robustness and feasibility of the approach.
- Combining local binary patterns and local color contrast for texture classification under varying illumination47
A texture descriptor for color texture classification specially designed to be robust against changes in the illumination conditions, which outperforms the original LBP approach and its color variants, even when these are computed after color normalization.
- Gappy PCA Classification for Occlusion Tolerant 3D Face Detection47
The proposed approach for the detection of faces in three dimensional scenes is tolerant against partial occlusions produced by the presence of any kind of object and can be used to improve the robustness of all those systems requiring a face detection stage in non-controlled scenarios.
- On the detection of pornographic digital images46
Both, decision forests of trees built according to CART (Classification And Regression Trees) methodology and Support Vectors Machines (SVM) have been used to perform the classification of pornographic and non-pornographic photographs.
- Image orientation detection using LBP-based features and logistic regression41
An algorithm for the automatic detection of the image orientation that relies on the image content as described by Local Binary Patterns (LBP) is presented and its accuracy is comparable with that of human observers in detecting the correct orientation of a wide range of image contents.
- TreEnhance: A tree search method for low-light image enhancement40
The method combines tree search theory, and in particular the Monte Carlo Tree Search (MCTS) algorithm, with deep reinforcement learning and proposes a guided search strategy that reverses the enhancement procedure that a photo editor applied to a given input image.
- 40
- Improving CNN-Based Texture Classification by Color Balancing40
It is shown how suitable color balancing models allow for a significant improvement in the accuracy in recognizing textures for many CNN architectures.
- AUTOMATIC CLASSIFICATION OF DIGITAL PHOTOGRAPHS BASED ON DECISION FORESTS37
It is shown here how low-level features can be related to semantic photo categories, such as indoor, outdoor and close-up, using decision forests consisting of trees constructed according to CART methodology.
- Combining multiple features for color texture classification34
It is shown that the use of these descriptors in a multiple classifiers framework makes it possible to achieve a very high classification accuracy in classifying texture images acquired under different lighting conditions and how the proposed combining strategy hand-crafted and convolutional neural networks features can be used together to further improve the classification accuracy.
- Visual recognition of aircraft mechanical parts for smart maintenance33
A visual recognition module of aircraft mechanical parts that has been included in a prototype system designed for the smart maintenance of the Alenia-Aermacchi M346 confirms the feasibility of the proposed recognition module in a very challenging and realistic condition.
- Halfway through the semantic gap: Prosemantic features for image retrieval33
An image description approach based on prosemantic features, which are obtained through a two-level feature extraction process and used to index images in an image retrieval system where searches are performed using relevance feedback.
- 32
- Remote Sensing Image Classification Exploiting Multiple Kernel Learning25
A novel procedure is presented that allows MKL to achieve good performance in the case of small training sets and experimental results on publicly available data sets demonstrate the feasibility of the proposed approach.
- Topic Modeling for Faster Literature Screening Using Transformer-Based Embeddings20
BERTopic, a transformer-based topic-modeling algorithm, was applied to two datasets used in recently published systematic reviews on peri-implantitis and bone regeneration, suggesting that adding a topic-modeling step to the screening process could potentially save working hours for researchers involved in systematic reviews of the literature.
- Learning Parametric Functions for Color Image Enhancement20
A novel CNN-based method for image enhancement that simulates an expert retoucher that is fast and accurate at the same time thanks to the decoupling between the inference of the parameters and the color transformation.
- Local Angular Patterns for Color Texture Classification18
A new color texture descriptor, specially designed to be robust to changes in the color of the illuminant, is defined, that is called Local Angular Patterns, which outperforms the state-of-the-art on a dataset of food textures.
- Illuminant Invariant Descriptors for Color Texture Classification17
A novel descriptor for color texture analysis specially designed to deal with changes in illumination in imaging is presented, based on a combination of the LBP approach with a measure of color contrast defined as the angle between two color vectors in an orthonormal color space.
- Intensity and color descriptors for texture classification16
This paper presents a descriptor for texture classification based on the histogram of a local measure of the color contrast that shows a performance improvement with respect to results achieved by baseline methods thus demonstrating the effectiveness of the proposed texture features.
- Color Target Localization under Varying Illumination Conditions16
The experimental results on a public dataset of RAW images containing the Macbeth ColorChecker CC target and acquired in uncontrolled environments, showed that all the descriptors considered benefited from the hypothesis validation introduced.
- T1K+: A Database for Benchmarking Color Texture Classification and Retrieval Methods15
A very large, heterogeneous database of high-quality texture images acquired under variable conditions, T1K+ contains 1129 classes of textures, which allows the design of experiments especially aimed at understanding the specific issues related to texture classification and retrieval.
- 15
- 15
- 14
- On the use of supervised features for unsupervised image categorization: An evaluation12
It is demonstrated that supervised features can be effectively used also for unsupervised image categorization, that is, for grouping semantically similar images.
- Prosemantic Features for Content-Based Image Retrieval12
An image description approach based on prosemantic features that allows for a more successful and quick retrieval of the query images and to verify the effectiveness of the approach, a target search experiment is designed.
- Content Aware Image Enhancement10
The overall quality of images is improved by a modular procedure automatically driven by the image class and content by integrating imaging and vision for content-aware enhancement and processing of digital photographs.
- Image Annotation for Adaptive Enhancement of Uncalibrated Color Images10
An innovative image annotation tool, based on a multi-class Support Vector Machine, for classifying image pixels in one of seven classes – sky, skin, vegetation, snow, water, ground, and man-made structures – or as unknown, is described.
- Offset equivariant networks and their applications9
This paper presents a framework for the design and implementation of offset equivariant networks, that is, neural networks that preserve in their output uniform increments in the input that achieves equivariance with respect to the photometric transformations that characterize changes in the lighting conditions.
- 9
- CURL: Image Classification using co-training and Unsupervised Representation Learning9
A strategy for semi-supervised image classification that leverages unsupervised representation learning and co-training, called CURL, that iteratively builds two classifiers on two different views of the data that clearly outperforms other supervised and semi- supervised learning methods in the state of the art.
- Color transfer using semantic image annotation9
An automatic local color transfer method based on semantic image annotation that permits to automatically transfer the color distribution from regions of the source and target images annotated with the same class (for example the class "sky").
- 8
- On the Robustness of Color Texture Descriptors across Illuminants8
This paper recovered the spectral reflectances of 1360 texture images from the Outex 13 data set and rendered them as if they were taken under 33 different illuminants, measuring the classification performance of the texture features considered.
- Classification-Based Color Constancy8
This work has designed a strategy for the selection of the most appropriate algorithm on the basis of the classification results, and tested it on a subset of the widely used Funt and Ciurea dataset.
- Exemplar-Free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation7
A new method for exemplar-free class incremental training of ViTs, which allows for plasticity towards new tasks while limiting forgetting of previous ones, and a new method of feature drift compensation that accommodates feature drift in the backbone when learning new tasks.
- CNN-Based Refactoring of Hand-Designed Filters for Texture Analysis: A Classic Revisited7
This work proposes the use of convolutional networks for refactoring traditional, hand-designed filters, and shows that the refactored filters can be conveniently used ‘off-the-shelf’ to achieve better performance than obtained with the original filters, but at the same computational cost.
- Image orientation detection using low-level features and faces7
This work proposes a system which automatically detects the correct orientation of digital photographs, which exploits the information provided by a face detector and a set of low-level features related to distributions in the image of color and edges.
- Content-Based Classification of Digital Photos7
It is shown that it is possible to relate low-level visual features to semantic photo categories, such as indoor, outdoor and close-up, using tree classifiers.
- Artistic Photo Filtering Recognition Using CNNs6
This paper proposes an approach based on deep Convolutional Neural Networks to recognize artistic photo filters applied to images, and shows that not only it is possible to reliably determine whether or not one of these filters has been applied, but also which one.
- 4
- Embeddings for Efficient Literature Screening: A Primer for Life Science Investigators4
The present work aims at introducing the technical evolution of the meaning representation systems, from vectors to embeddings and transformers tailored to life science investigators with no previous knowledge of natural language processing.
- 4
- Prosemantic Image Retrieval4
The system effectiveness will be proved for both category and target search on two standard image databases, even without a "good" initial example and ancillary information, such as device metadata, text annotations, etc.
- Color Constancy Algorithm Selection Using CART4
This work investigates how illuminant estimation techniques can be improved taking into account intrinsic, low level properties of the images, and shows how these properties can be used to drive the selection of the best algorithm for a given image.
- Semantic classification, low level features and relevance feedback for content-based image retrieval4
The qualitative experiments performed on a database of more than 46,000 photos downloaded from the Web show that the combination of semantic and low level features coupled with a relevance feedback algorithm, effectively improve the accuracy of the image retrieval sessions.
- 4
- Semantic labeling of digital photos by classification4
This paper addresses the problem of annotating photographs with broad semantic labels by designing a hierarchical classification strategy which first classifies images as pornographic or not-pornographic, and then classification as indoor, outdoor, or close-up.
- 3
- Explaining image enhancement black-box methods through a path planning based algorithm2
A path planning algorithm which provides a step-by-step explanation of the output produced by state-of-the-art enhancement methods, and obtaining sequences of enhancing operators able to produce very similar results in terms of performance and overcoming the huge limitation of poor interpretability of the best performing algorithms.
- With a little help from my friends2
A content-based method for semi-automatic organization of photo albums based on the analysis of how different users organize their own pictures to help the user in dividing his pictures into groups characterized by a similar semantic content.
- Region-Based Annotation of Digital Photographs2
A region-based method for the annotation of outdoor photographs using the normalized cut algorithm and a rejection option is applied to discard those regions for which the classifier is not confident enough.
- Occlusions in Face Recognition: a 3D Approach2
A method for removing glasses from a frontal image of the human face using a cascade of classifiers trained to detect subparts of the face and a method using local approaches which divide the face into parts which are independently compared.
- No-reference metrics for JPEG: analysis and refinement using wavelets2
Experimental results show that information about the visual content of the image can be exploited to improve the estimation of the quality of JPEG compressed images.
- INTEGRATING IMAGING AND VISION FOR CONTENT-SPECIFIC IMAGE ENHANCEMENT2
The overall quality of digital photographs is improved by a modular, image enhancement procedure driven by the image content, which can be combined to improve the overall quality according to image and defect categories.
- A permutation-equivariant deep learning model for quantum state characterization1
This paper studies how to combine a permutation-equivariant deep learning model with the tQST protocol, and tests the model on quantum state tomography and purity estimation and validate the robustness of the model to noise.
- 1
- 1
- On the Importance of Domain Adaptation in Texture Classification1
This paper is the first study investigating up to which extent this visual recognition problem suffers from this issue, and the effectiveness of existing domain adaptation algorithms in mitigating it, and focuses on domain adaptation methods based on shallow classifiers.
- Searching through photographic databases with QuickLook1
The prosemantic features are obtained by a two-step feature extraction process and are used as input to a bank of classifiers, each one trained to recognize a given semantic category, to produce score vectors.
- Unsupervised Classemes1
A new model of semantic features is presented that, unlike previously presented methods, does not rely on the presence of a labeled training data base, as the creation of the feature extraction function is done in an unsupervised manner.
- Spatially organized visualization of image query results1
This work presents a system which visualizes the results obtained from image search engines in such a way that users can conveniently browse the retrieved images.
- Automatic annotation of outdoor photographs1
A strategy for the automatic annotation of outdoor photographs using a normalized cut segmentation to identify the regions to be classified by a multi-class Support Vector Machine.
- On the Coöperative Creation of Multimedia Meaning1
A content-based method for semi- automatic organization of photo albums based on the analysis of how different users organize their own pictures to help the user in dividing his pictures into groups characterized by a similar semantic content.
- FaceLab: a tool for performance evaluation of face recognition strategies1
FaceLab is presented, an innovative, open environment created to evaluate the performance of face recognition strategies, which simplifies the basic steps involved in testing procedures such as data organization and preprocessing, definition and management of training and test sets, definition of recognition strategies and automatic computation of performance measures.
- Tri-dimensional face detection and localization1
An innovative method that combines a feature-based approach with a holistic approach for tri-dimensional face detection and localization and has been tested on about 150 3D faces acquired by a laser range scanner with good results.
- –
- –
- –
- –
- –
- Special Section Guest Editorial:Color in Texture and Material Recognition–
This PDF file contains the editorial “Special Section Guest Editorial: color in Texture and Material Recognition” for JEI Vol.
- –
- –
- –
- Image Indexing Using Prosemantic Features–
Target search experiments show that the use of prosemantic features, combined with the relevance feedback mechanism of QuickLook, allows for a more successful and quick retrieval of the query images with respect to low-level features.
- High-Level Features for Image Indexing and Retrieval–
Image indexing for content-based image retrieval is the process of automatically computing a compact representation (numerical or alphanumerical) of some attribute of digital images, to be used to derive information about the image contents.
- –
- –
- Cooperative classification of shared images–
A quantitative evaluation of the proposed approach is obtained by simulating the amount of user interaction needed to annotate the albums of a set of members of the flickr R(trademark) photo-sharing community.
- –
- –
- –
- Exploiting CART methodology in digital color imaging: summary of an experience–
The experience in the use of CART methodology for the automatic semantic classification of digital images based on low-level features describing some of their pictorial characteristics is summarized.
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.