Publications

Object-centric models inspired by DETR have become the dominant paradigm for open-vocabulary video instance segmentation (OV-VIS). While recent efforts have reduced the computational cost of pixel decoding, textual modality fusion, and object decoding to make these architectures more suitable for mobile devices, real-time on-device inference at high frame rates remains an open challenge. In this paper, we introduce SegFS, a dual-stream fast-slow framework that significantly improves efficiency without sacrificing accuracy. On sparse keyframes, an open-vocabulary object-based model predicts instance-level representations. These representations are then projected back into the backbone feature space to condition a lightweight fast network, which efficiently relocalizes and segments the instances in subsequent frames. By shifting instance propagation from object decoding to feature-space conditioning, our approach decouples multimodal semantic understanding from dense mask prediction and enables efficient temporal propagation. The proposed fast branch achieves up to 14x lower latency than the mobile-oriented MOBIUS model, while maintaining competitive segmentation performance on standard OV-VIS benchmarks.

Wrist-worn wearables are the dominant paradigm for step counting, but their performance degrades noticeably when used by people with motor impairments. We instrument a standard cane with a low-cost sensing framework to address this and, at the same time, to lay the foundation for a self-supervised gait-quality assessment in which a convolutional autoencoder trained only on normal walking uses reconstruction error as an anomaly score. A downward-facing time-of-flight sensor measures the distance of the cane tip from the ground and provides the reference against which step counting is validated, reaching 92.9% accuracy, while on the same recordings the anomaly detector reaches 98.7% AUPRC under session-level cross-validation, also ordering the types of gait problems by how strongly they deform the signal.

Accurate delineation of maxillofacial anatomy in Cone-Beam Computed Tomography (CBCT) is essential for dental planning, but robust automated segmentation remains challenging, due to limited public multi-structure datasets and the high computational burden of 3D deep learning models. We present and release ToothFairy3, a large-scale CBCT benchmark that extends ToothFairy2 with 102 additional fully annotated scans and an expanded taxonomy covering 77 classes, including 32 tooth-specific pulp cavities and small neurovascular structures. ToothFairy3 comprises 582 volumes (over 40000 annotated objects), with 532 released with voxel-level labels and 50 held out for leakage-free, server-side evaluation. We also introduce U-Mamba2, an efficient U-Net-style architecture that inserts a Mamba2 state-space block at the bottleneck to capture global context with favorable computational scaling. Our proposed domain-informed training further improves the learning of maxillofacial anatomies. Across CNN, Transformer, and Mamba baselines, U-Mamba2 achieves competitive Dice/HD95 scores with lower latency and, compared with training on state-of-the-art public CBCT datasets, ToothFairy3-trained models generalize best to the hidden test set, particularly for maxillary structures.

Automated plant species identification from citizen-science imagery is an established, demanding fine-grained recognition problem: large taxonomic label spaces, visually similar species, and long-tailed observations require real model capacity, while field use constrains memory, latency, and power. Model size is only part of the deployment cost: intermediate activations held in memory during inference and platformdependent execution behavior matter too, so compact recognition must be assessed on target hardware rather than through complexity metrics alone. We present BoltNet, an ultra-lightweight fully convolutional architecture combining a Spatial Redistribution Bottleneck and Logit PreSampling to improve the tradeoff between predictive performance and model size in high-cardinality classification, and report the AccuracyCompression Tradeoff as a complementary diagnostic. On Pl@ntNet300K, BoltNet reaches 0.682 F1-score with 341K parameters (1.37 MB), the highest F1-score among evaluated models below 2 MB and close to substantially larger convolutional backbones. Model-only measurements on a Raspberry Pi 5, Jetson Orin Nano, and Hailo-8 characterize execution across CPU, GPU, and NPU platforms, where BoltNet is the most consistently efficient model, with the best FPS/W on the GPU and NPU and second-best on the CPU. Results on AIDERv2 and CLRS provide secondary evidence of transfer across environmental image-classification tasks. Code available at: https://codeberg.org/danielrossi/BoltNet.

Adapting large pre-trained models to downstream tasks often produces task-specific parameter updates that are expensive to relearn for every model variant. While recent work has shown that such updates can be transferred between models with identical architectures, transferring them across models of different widths remains unexplored. In this work, we introduce Theseus, a training-free method for transporting task updates across heterogeneous-width models. Rather than matching parameters, we characterize a task update by the functional effect it induces on intermediate representations. We formalize task-vector transport as a functional matching problem on observed activations and show that, after aligning representation spaces via orthogonal Procrustes analysis, it admits a stable closed-form solution that preserves the geometry of the update. We evaluate Theseus on vision and language models across different widths, showing consistent improvements over baselines without additional training or backpropagation. Our results show that task updates can be meaningfully transferred across architectures when task identity is defined functionally rather than parametrically. Code is available at https://github.com/apanariello4/merge-and-rebase.

This study presents an interdisciplinary methodology for detecting biblical references in Latin patristic literature through an innovative combination of rigorous philological approach and Natural Language Processing (NLP) techniques. Focusing on one of the most influential ancient Christian commentaries on the Bible, Augustine of Hippo’s De Genesi ad litteram, and its relationship with Latin biblical texts (specifically, Jerome’s Vulgate and pre-Vulgate versions), this research introduces a token-based classification system for intertextual references, enriched with semantic annotations and supported by the INCEpTION platform. The first section shows how this numerical classification system accounts for exact matches, lemmatized forms, roots, synonyms, and other forms of semantic parallels (here referred to as “structures”), capturing a wide spectrum of textual similarity. To enhance automatic retrieval of these intertextual connections, we fine-tune BERT-based language models for Latin, incorporating contrastive learning and hard negative mining. In the second section, experimental results show that finetuned models significantly outperform baseline models at various levels of textual similarity. This work highlights the utility of computational models in overcoming the traditional dichotomy between explicit quotations and implicit allusions, embracing multiple intermediate nuances of similarity and offering a scalable approach to the study of intertextuality in ancient writings.

Fashion illustration is a crucial medium for designers to convey their creative vision and transform design concepts into tangible representations that showcase the interplay between clothing and the human body. In the context of fashion design, computer vision techniques have the potential to enhance and streamline the design process. Departing from prior research primarily focused on virtual try-on, this paper tackles the task of multimodal-conditioned fashion image editing. Our approach aims to generate human-centric fashion images guided by multimodal prompts, including text, human body poses, garment sketches, and fabric textures. To address this problem, we propose extending latent diffusion models to incorporate these multiple modalities and modifying the structure of the denoising network, taking multimodal prompts as input. To condition the proposed architecture on fabric textures, we employ textual inversion techniques and let diverse cross-attention layers of the denoising network attend to textual and texture information, thus incorporating different granularity conditioning details. Given the lack of datasets for the task, we extend two existing fashion datasets, Dress Code and VITON-HD, with multimodal annotations. Experimental evaluations demonstrate the effectiveness of our proposed approach in terms of realism and coherence concerning the provided multimodal inputs.