Thomas A. Ciarfuglia — accepted AI-conference papers

2025

MAD-BA: 3D LiDAR Bundle Adjustment - From Uncertainty Modelling to Structure Optimization

Krzysztof Cwian, Luca Di Giammarino, Simone Ferrari, Thomas A. Ciarfuglia, Giorgio Grisetti, Piotr Skrzypczynski

RA-L 2025

The joint optimization of sensor poses and 3D structure is fundamental for state estimation in robotics and related fields. Current LiDAR systems often prioritize pose optimization, with structure refinement either omitted or treated separately using implicit representations. This paper introduces a

Cited by 3 Source ↗Scholar ↗

2024

AgriSORT: A Simple Online Real-time Tracking-by-Detection framework for robotics in precision agriculture

Leonardo Saraceni, Ionut M. Motoi, Daniele Nardi, Thomas A. Ciarfuglia

ICRA 2024poster

The problem of multi-object tracking (MOT) consists in detecting and tracking all the objects in a video sequence while keeping a unique identifier for each object. It is a challenging and fundamental problem for robotics. In precision agriculture the challenge of achieving a satisfactory solution i…

Cited by 10 Source ↗code ↗Scholar ↗

2019

Weakly Supervised Fruit Counting for Yield Estimation Using Spatial Consistency

Enrico Bellocchio, Thomas A. Ciarfuglia, Gabriele Costante, Paolo Valigi

RA-L 2019

Fruit counting is a fundamental component for yield estimation applications. Most of the existing approaches address this problem by relying on fruit models (i.e., by using object detectors) or by explicitly learning to count. Despite the impressive results achieved by these approaches, all of them

Cited by 50 Source ↗Scholar ↗

2018

Full-GRU Natural Language Video Description for Service Robotics Applications

Silvia Cascianelli, Gabriele Costante, Thomas A. Ciarfuglia, Paolo Valigi, Mario Luca Fravolini

RA-L 2018

Enabling effective human-robot interaction is crucial for any service robotics application. In this context, a fundamental aspect is the development of a user-friendly human-robot interface, such as a natural language interface. In this letter, we investigate the robot side of the interface, in part

Cited by 31 Source ↗Scholar ↗

2018

J-MOD2: Joint Monocular Obstacle Detection and Depth Estimation

Michele Mancini, Gabriele Costante, Paolo Valigi, Thomas A. Ciarfuglia

RA-L 2018

In this letter, we propose an end-to-end deep architecture that jointly learns to detect obstacles and estimate their depth for MAV flight applications. Most of the existing approaches rely either on Visual simultaneous localization and mapping (SLAM) systems or on depth estimation models to build t

Cited by 53 Source ↗Scholar ↗

2018

LS-VO: Learning Dense Optical Subspace for Robust Visual Odometry Estimation

Gabriele Costante, Thomas A. Ciarfuglia

RA-L 2018

This work proposes a novel deep network architecture to solve the camera ego-motion estimation problem. A motion estimation network generally learns features similar to optical flow (OF) fields starting from sequences of images. This OF can be described by a lower dimensional latent space. Previous

Cited by 76 Source ↗Scholar ↗

2017

Toward Domain Independence for Learning-Based Monocular Depth Estimation

Michele Mancini, Gabriele Costante, Paolo Valigi, Thomas A. Ciarfuglia, Jeffrey A. Delmerico, Davide Scaramuzza

RA-L 2017

Modern autonomous mobile robots require a strong understanding of their surroundings in order to safely operate in cluttered and dynamic environments. Monocular depth estimation offers a geometry-independent paradigm to detect free, navigable space with minimum space, and power consumption. These re

Cited by 65 Source ↗Scholar ↗

2016

Exploring Representation Learning With CNNs for Frame-to-Frame Ego-Motion Estimation

Gabriele Costante, Michele Mancini, Paolo Valigi, Thomas A. Ciarfuglia

RA-L 2016

Visual ego-motion estimation, or briefly visual odometry (VO), is one of the key building blocks of modern SLAM systems. In the last decade, impressive results have been demonstrated in the context of visual navigation, reaching very high localization performance. However, all ego-motion estimation

Cited by 204 Source ↗Scholar ↗

2016

Fast robust monocular depth estimation for Obstacle Detection with fully convolutional networks

Michele Mancini, Gabriele Costante, Paolo Valigi, Thomas A. Ciarfuglia

IROS 2016poster

Obstacle Detection is a central problem for any robotic system, and critical for autonomous systems that travel at high speeds in unpredictable environment. This is often achieved through scene depth estimation, by various means. When fast motion is considered, the detection range must be longer eno…

Cited by 142 Source ↗Scholar ↗