Search Results for author:
Found 60 papers, 26 papers with code
Leveraging Acoustic Images for Effective Self-Supervised Audio Representation Learning
However, since 2D planar arrays are cumbersome and not as widespread as ordinary microphones, we propose that the richer information content of acoustic images can be distilled, through a self-supervised learning scheme, into more powerful audio and visual feature representations.
Dif4FF: Leveraging Multimodal Diffusion Models and Graph Neural Networks for Accurate New Fashion Product Performance Forecasting
As a result, this paper proposes Dif4FF, a novel two-stage pipeline for New Fashion Product Performance Forecasting (NFPPF) that leverages the power of diffusion models conditioned on multimodal data related to specific clothes.
Pre-trained Multiple Latent Variable Generative Models are good defenders against Adversarial Attacks
Attackers can deliberately perturb classifiers' input with subtle noise, altering final predictions.
Collaborative Instance Navigation: Leveraging Agent Self-Dialogue to Minimize User Input
Existing embodied instance goal navigation tasks, driven by natural language, assume human users to provide complete and nuanced instance descriptions prior to the navigation, which can be impractical in the real world as human instructions might be brief and ambiguous.
Material synthesis through simulations guided by machine learning: a position paper
In this position paper, we propose an approach for sustainable data collection in the field of optimal mix design for marble sludge reuse.
IoT-Based Coma Patient Monitoring System
Continuous monitoring of coma patients is essential but challenging, especially in developing countries with limited resources, staff, and infrastructure.
Learning based Ge'ez character handwritten recognition
Ge'ez, an ancient Ethiopic script of cultural and historical significance, has been largely neglected in handwriting recognition research, hindering the digitization of valuable manuscripts.
Upper-Body Pose-based Gaze Estimation for Privacy-Preserving 3D Gaze Target Detection
This paper presents a novel approach to tackle this problem by utilizing the person's upper-body pose and available depth maps to extract a 3D gaze direction and employing a multi-stage or an end-to-end pipeline to predict the gazed target.
Multi-Camera Industrial Open-Set Person Re-Identification and Tracking
This work presents MICRO-TRACK, a Modular Industrial multi-Camera Re_identification and Open-set Tracking system that is real-time, scalable, and easy to integrate into existing industrial surveillance scenarios.
SITUATE: Indoor Human Trajectory Prediction through Geometric Features and Self-Supervised Vision Representation
Patterns of human motion in outdoor and indoor environments are substantially different due to the scope of the environment and the typical intentions of people therein.
Self-Supervised Iterative Refinement for Anomaly Detection in Industrial Quality Control
This study introduces the Iterative Refinement Process (IRP), a robust anomaly detection methodology designed for high-stakes industrial quality control.
Enhancing Split Computing and Early Exit Applications through Predefined Sparsity
This paper introduces a novel approach to address this challenge by combining the concept of predefined sparsity with Split Computing (SC) and Early Exit (EE).
MTL-Split: Multi-Task Learning for Edge Devices using Split Computing
However, how to partition such a multi-tasking DNN to be deployed within a SC framework has not been sufficiently studied.
Leveraging Latent Diffusion Models for Training-Free In-Distribution Data Augmentation for Surface Defect Detection
Usually, defect detection classifiers are trained on ground-truth data formed by normal samples (negative data) and samples with defects (positive data), where the latter are consistently fewer than normal samples.
I2EDL: Interactive Instruction Error Detection and Localization
We evaluate the proposed I2EDL on a dataset of instructions containing errors, and further devise a novel metric, the Success weighted by Interaction Number (SIN), to reflect both the navigation performance and the interaction effectiveness.
Diffusion-based Image Generation for In-distribution Data Augmentation in Surface Defect Detection
In general, defect detection classifiers are trained on ground-truth data formed by normal samples (negative data) and samples with defects (positive data), where the latter are consistently fewer than normal samples.
Exploring 3D Human Pose Estimation and Forecasting from the Robot's Perspective: The HARPER Dataset
The scenario underlying HARPER includes 15 actions, of which 10 involve physical contact between the robot and users.
Mind the Error! Detection and Localization of Instruction Errors in Vision-and-Language Navigation
Moreover, we formally define the task of Instruction Error Detection and Localization, and establish an evaluation protocol on top of our benchmark dataset.
Disentangled Latent Spaces Facilitate Data-Driven Auxiliary Learning
In this paper, we propose a novel framework, dubbed Detaux, whereby a weakly supervised disentanglement procedure is used to discover a new unrelated auxiliary classification task, which allows us to go from a Single-Task Learning (STL) to a Multi-Task Learning (MTL) problem.
Graph-level Representation Learning with Joint-Embedding Predictive Architectures
They aim to learn an energy-based model by predicting the latent representation of a target signal y from the latent representation of a context signal x. JEPAs bypass the need for negative and positive samples, traditionally required by contrastive learning while avoiding the overfitting issues associated with generative pretraining.
Ranked #11 on
Graph Classification
on REDDIT-B
A Machine Learning-oriented Survey on Tiny Machine Learning
The emergence of Tiny Machine Learning (TinyML) has positively revolutionized the field of Artificial Intelligence by promoting the joint design of resource-constrained IoT hardware devices and their learning-based software architectures.
Language-enhanced RNR-Map: Querying Renderable Neural Radiance Field maps with natural language
We present Le-RNR-Map, a Language-enhanced Renderable Neural Radiance map for Visual Navigation with natural language query prompts.
Neuro-symbolic Empowered Denoising Diffusion Probabilistic Models for Real-time Anomaly Detection in Industry 4.0
Industry 4. 0 involves the integration of digital technologies, such as IoT, Big Data, and AI, into manufacturing and industrial processes to increase efficiency and productivity.
OO-dMVMT: A Deep Multi-view Multi-task Classification Framework for Real-time 3D Hand Gesture Classification and Segmentation
Continuous mid-air hand gesture recognition based on captured hand pose streams is fundamental for human-computer interaction, particularly in AR / VR.
Split-Et-Impera: A Framework for the Design of Distributed Deep Learning Applications
Many recent pattern recognition applications rely on complex distributed architectures in which sensing and computational nodes interact together through a communication network.
A Masked Face Classification Benchmark on Low-Resolution Surveillance Images
Results show that the richness of SF-MASK (real + synthetic images) leads all of the tested classifiers to perform better than exploiting comparative face mask datasets, on a fixed 1077 images testing set.
On the use of learning-based forecasting methods for ameliorating fashion business processes: A position paper
The fashion industry is one of the most active and competitive markets in the world, manufacturing millions of products and reaching large audiences every year.
Leveraging commonsense for object localisation in partial scenes
With the proposed graph-based scene representation, we estimate the unknown position of the target object using a Graph Neural Network that implements a novel attentional message passing mechanism.
Toward Smart Doors: A Position Paper
In particular, a smart door system predicts the intention of people near the door based on the social context of the surrounding environment and then makes rational decisions about whether or not to open the door.
I-SPLIT: Deep Network Interpretability for Split Computing
A neuron is important if its gradient with respect to the correct class decision is high.
Pose Forecasting in Industrial Human-Robot Collaboration
Pushing back the frontiers of collaborative robots in industrial environments, we propose a new Separable-Sparse Graph Convolutional Network (SeS-GCN) for pose forecasting.
POP: Mining POtential Performance of new fashion products via webly cross-modal query expansion
We propose a data-centric pipeline able to generate exogenous observation data for the New Fashion Product Performance Forecasting (NFPPF) problem, i. e., predicting the performance of a brand-new clothing probe with no available past observations.
Ranked #1 on
New Product Sales Forecasting
on VISUELLE
SHREC 2022 Track on Online Detection of Heterogeneous Gestures
This paper presents the outcomes of a contest organized to evaluate methods for the online recognition of heterogeneous gestures from sequences of 3D hand poses.
The multi-modal universe of fast-fashion: the Visuelle 2.0 benchmark
SO-fore assumes that the season has started and a set of new products is on the shelves of the different stores.
Short-observation new product sales forecasting
Time Series Analysis
Under the Hood of Transformer Networks for Trajectory Forecasting
This paper proposes the first in-depth study of Transformer Networks (TF) and Bidirectional Transformers (BERT) for the forecasting of the individual motion of people, without bells and whistles.
Spatial Commonsense Graph for Object Localisation in Partial Scenes
The SCG is used to estimate the unknown position of the target object in two steps: first, we feed the SCG into a novel Proximity Prediction Network, a graph neural network that uses attention to perform distance prediction between the node representing the target object and the nodes representing the observed objects in the SCG; second, we propose a Localisation Module based on circular intersection to estimate the object position using all the predicted pairwise distances in order to be independent of any reference system.
MovingFashion: a Benchmark for the Video-to-Shop Challenge
Retrieving clothes which are worn in social media videos (Instagram, TikTok) is the latest frontier of e-fashion, referred to as "video-to-shop" in the computer vision literature.
Ranked #1 on
Video-to-Shop
on MovingFashion
Well Googled is Half Done: Multimodal Forecasting of New Fashion Product Sales with Image-based Google Trends
In particular, we propose a neural network-based approach, where an encoder learns a representation of the exogenous time series, while the decoder forecasts the sales based on the Google Trends encoding and the available visual and metadata information.
Ranked #3 on
New Product Sales Forecasting
on VISUELLE
POMP: Pomcp-based Online Motion Planning for active visual search in indoor environments
Our POMP method uses as input the current pose of an agent (e. g. a robot) and a RGB-D frame.
Infinite Feature Selection: A Graph-based Feature Filtering Approach
We propose a filtering feature selection framework that considers subsets of features as paths in a graph, where a node is a feature and an edge indicates pairwise (customizable) relations among features, dealing with relevance and redundancy principles.
The Visual Social Distancing Problem
One of the main and most effective measures to contain the recent viral outbreak is the maintenance of the so-called Social Distancing (SD).
An integrated light management system with real-time light measurement and human perception
Illumination is important for well-being, productivity and safety across several environments, including offices, retail shops and industrial warehouses.
Transformer Networks for Trajectory Forecasting
In particular, the TF model without bells and whistles yields the best score on the largest and most challenging trajectory forecasting benchmark of TrajNet.
Ranked #12 on
Trajectory Prediction
on ETH/UCY
Texture Retrieval in the Wild through detection-based attributes
Capturing the essence of a textile image in a robust way is important to retrieve it in a large repository, especially if it has been acquired in the wild (by taking a photo of the textile of interest).
Texel-Att: Representing and Classifying Element-based Textures by Attributes
It first individuates texels, characterizing them with individual attributes; subsequently, texels are grouped and characterized through layout attributes, which give the Texel-Att representation.
SIMCO: SIMilarity-based object COunting
We present SIMCO, the first agnostic multi-class object counting approach.
Human-centric light sensing and estimation from RGBD images: The invisible light switch
ILS may therefore dim those luminaires, which are not seen by the user, resulting in an effective energy saving, especially in large open offices (where light may otherwise be ON everywhere for a single person).
Forecasting People Trajectories and Head Poses by Jointly Reasoning on Tracklets and Vislets
In this work, we explore the correlation between people trajectories and their head orientations.
RGBD2lux: Dense light intensity estimation with an RGBD sensor
The proposed method uses both depth data and images from the sensor to provide a dense measure of light intensity in the field of view of the camera.
MX-LSTM: mixing tracklets and vislets to jointly forecast trajectories and head poses
Recent approaches on trajectory forecasting use tracklets to predict the future positions of pedestrians exploiting Long Short Term Memory (LSTM) architectures.
“Seeing is Believing”: Pedestrian Trajectory Forecasting Using Visual Frustum of Attention
In this paper we show the importance of the head pose estimation in the task of trajectory forecasting.
Understanding Deep Architectures by Visual Summaries
In deep learning, visualization techniques extract the salient patterns exploited by deep networks for image classification, focusing on single images; no effort has been spent in investigating whether these patterns are systematically related to precise semantic entities over multiple images belonging to a same class, thus failing to capture the very understanding of the image class the network has realized.
What your Facebook Profile Picture Reveals about your Personality
We analyze the effectiveness of four families of visual features and we discuss some human interpretable patterns that explain the personality traits of the individuals.
Clothing and People - A Social Signal Processing Perspective
In our society and century, clothing is not anymore used only as a means for body protection.
Looking Beyond Appearances: Synthetic Training Data for Deep CNNs in Re-identification
Re-identification is generally carried out by encoding the appearance of a subject in terms of outfit, suggesting scenarios where people do not change their attire.
Infinite Feature Selection
Filter-based feature selection has become crucial in many classification settings, especially object recognition, recently faced with feature learning strategies that originate thousands of cues.
The S-Hock Dataset: Analyzing Crowds at the Stadium
In the dataset, a massive annotation has been carried out, focusing on the spectators at different levels of details: at a higher level, people have been labeled depending on the team they are supporting and the fact that they know the people close to them; going to the lower levels, standard pose information has been considered (regarding the head, the body) but also fine grained actions such as hands on hips, clapping hands etc.
Audio Surveillance: a Systematic Review
Despite surveillance systems are becoming increasingly ubiquitous in our living environment, automated surveillance, currently based on video sensory modality and machine intelligence, lacks most of the time the robustness and reliability required in several real applications.
F-formation Detection: Individuating Free-standing Conversational Groups in Images
Then, as a main contribution, we present a brand new method for the automatic detection of groups in still images, which is based on a graph-cuts framework for clustering individuals; in particular we are able to codify in a computational sense the sociological definition of F-formation, that is very useful to encode a group having only proxemic information: position and orientation of people.
Free energy score space
Score functions induced by generative models extract fixed-dimension feature vectors from different-length data observations by subsuming the process of data generation, projecting them in highly informative spaces called score spaces.
