A5042846340- Face recognition and analysis
- Video Surveillance and Tracking Methods
- Digital Media Forensic Detection
- 3D Surveying and Cultural Heritage
- Advanced Image and Video Retrieval Techniques
- Digital and Cyber Forensics
- Anomaly Detection Techniques and Applications
- Image Retrieval and Classification Techniques
- Biometric Identification and Security
- Image Processing and 3D Reconstruction
- Face and Expression Recognition
- Advanced Neural Network Applications
- E-commerce and Technology Innovations
- Medical Image Segmentation Techniques
- Advanced Steganography and Watermarking Techniques
- Network Security and Intrusion Detection
- Video Analysis and Summarization
- Generative Adversarial Networks and Image Synthesis
- Archaeological Research and Protection
- Remote Sensing and LiDAR Applications
- Advanced Malware Detection Techniques
- Fire effects on ecosystems
- Gait Recognition and Analysis
- Advanced Vision and Imaging
- Advanced Data Compression Techniques
Universidade Estadual de Campinas (UNICAMP)
2012-2024
Institute of Electrical and Electronics Engineers
2020
Regional Municipality of Niagara
2020
IEEE Computer Society
2020
Samsung (Brazil)
2015-2016
University of California, San Diego
2015
Pontifícia Universidade Católica do Paraná
2015
Universidade Federal do Paraná
2015
In recent years, software-based face presentation attack detection (PAD) methods have seen a great progress. However, most existing schemes are not able to generalize well in more realistic conditions. The objective of this competition is evaluate and compare the generalization performances mobile PAD techniques under some real-world variations, including unseen input sensors, instruments (PAI) illumination conditions, on larger scale OULU-NPU dataset using its standard evaluation protocols...
Learning from fully-unlabeled data is challenging in Multimedia Forensics problems, such as Person Re-Identification and Text Authorship Attribution. Recent self-supervised learning methods have shown to be effective when dealing with cases where the underlying classes significant semantic differences, intra-class distances are substantially lower than inter-class distances. However, this not case for forensic applications which similar semantics training test sets disjoint identities....
With the widespread use of biometric authentication comes exploitation presentation attacks, possibly undermining effectiveness these technologies in real-world setups. One example takes place when an impostor, aiming at unlocking someone else's smartphone, deceives built-in face recognition system by presenting a printed image user. In this work, we study problem automatically detecting attacks against methods, considering use-case fast device and hardware constraints mobile devices. To...
In this article we present the first effective method based on global optimization for reconstruction of image puzzles comprising rectangle pieces-Puzzle Solving by Quadratic Programming (PSQP). The proposed novel mathematical formulation reduces problem to maximization a constrained quadratic function, which is solved via gradient ascent approach. deterministic and can deal with arbitrary identical rectangular pieces. We provide experimental results showing its effectiveness when compared...
We present a new formulation to automatically solve jigsaw puzzles considering only the information contained on image. Our maps problem of solving puzzle maximization constrained quadratic function that can be solved by numerical method. The proposed method is deterministic and it handle arbitrary rectangular pieces. tested validity problems up 3300 pieces, we compared our results current state-of-the-art, obtaining superior accuracy.
Person Re-Identification (ReID) across non-overlapping cameras is a challenging task, and most works in prior art rely on supervised feature learning from labeled dataset to match the same person different views. However, it demands time-consuming task of labeling acquired data, prohibiting its fast deployment forensic scenarios. Unsupervised Domain Adaptation (UDA) emerges as promising alternative, performs adaptation model trained source target domain without identity-label annotation....
Considering that blood fluoride concentration varies according to exposure and dental fluorosis is related the amount of enamel formed under a given dose, present study investigated whether produced by an oscillating chronic dose would be similar caused constant representing mean oscillation during time. Rats received 78 days water with concentrations 0, 12.5, 25 or 37.5 microg F/mL, 12.5 F/mL every 72 h (mean exposure=25 F/mL). The in plasma, femur incisors rats were determined at end...
The recovery of footwear impression in crime scenes plays an important role investigations to corroborate or refute information, narrow down the number suspects. Casting 3D impressions is a long-standing standard obtain models prints, slowly being replaced by less invasive method, scanning. In this paper, we present alternative method based on multiview stereo that yields accurate model and provides some benefits over existing methods. We evaluate results comparing our reconstructed with...
One of the major modern threats to society is propagation misinformation — fake news, science denialism, hate speech fueled by social media's widespread adoption. On leading platforms, millions automated and profiles exist only for this purpose. step mitigate problem verifying authenticity profiles, which proves be an infeasible task done manually. Recent data-driven methods accurately tackle performing automatic authorship attribution, although important aspect often overlooked: model...
Most pictures shared online are accompanied by temporal metadata (i.e., the day and time they were taken), which makes it possible to associate an image content with real-world events. Maliciously manipulating this can convey a distorted version of reality. In work, we present emerging problem detecting timestamp manipulation. We propose end-to-end approach verify whether purported capture outdoor is consistent its geographic location. consider manipulations done in hour and/or month...
Tensor Scale is a morphometric parameter that unifies the representation of local structure thickness, orientation, and anisotropy, which can be used in several image processing tasks. This paper introduces new application for tensor scale, detection saliences on given contour, based scale orientations computed entire object mapped to its contour. For validation purposes, we present shape descriptor uses detected contour saliences. Experimental results are provided, comparing proposed method...
Fake news has been certainly the expression of moment: from political round table discussions to newspapers social and mainstream media. It is everywhere. With such an intense discussion yet few effective ways combat it, what can be done? Providing methods fight back even least harming hoax a responsibility. To look for authenticity in wide sea fake news, every detail lead. Image appearance semantic content text images are some main properties, which analyzed reveal slightest lie. In this...
We present a simple yet effective framework - Transmitting What Matters (TWM) to generate compressed videos containing only relevant objects targeted specific computer vision tasks, such as faces for the task of face expression recognition, license plates optical character among others. TWM takes advantage final desired compose video frames with necessary data. The are and can be stored or transmitted powerful servers where extensive time-consuming tasks performed. experimentally trade-offs...
We live in a connected society which no major event - from music concerts to terrorist attempts happens without being recorded by smartphone and shared instantly the world. This flow of generated data is often unstructured carries reliable information with respect time capture such media pieces. Consequently, posterior reconstruction, understanding fact-checking that are hindered if not properly organized. In this work, we train data-driven method chronologically sort images originated real...
We discuss the problem of restructuring visual data from different heterogeneous sources to analyze an event interest. present X-coherence: a pipeline seeking organize and represent pieces data, tying them coherently with real world one another. also outline research challenges while X-coherence.
In this paper, we describe how an effective vanishing point detector can be applied to photogrammetry when only a single view of architectural environment is available. Our method performs automatic segment clustering in projective space - direct transformation from the image instead traditional bounded accumulator space. Experiments on real images show effectiveness proposed finding all points, as well its application algorithm, by recovering vertical direction scene and line for ground plane.
Mobile devices have their popularity and affordability greatly increased in recent years. As a consequence of ubiquity, these now carry all sorts personal data that should be accessed only by owner. Even though knowledge-based procedures are still the main methods to secure owner's identity, recently biometric traits been employed for more effortless authentication. In this work, authors propose facial verification method optimised mobile environment. It consists two-tiered procedure...