arXiv subjects

Andrei Arhire

Publications and source records attributed to Andrei Arhire.

5 recordsLinked to original sources

arXiv2026

LightMIS: Ultra-Lightweight Medical Image Segmentation Without a Stage-Wise Decoder

We present LightMIS, a scalable family of ultra-lightweight convolutional networks for 2D binary medical image segmentation without a learned stage-wise decoder. LightMIS aligns the outputs of a five-level encoder to a common resolution using Scale-Aligned Projection blocks, aggregates them once, and refines the fused representation with an Adaptive Fusion Cascade. The cascade combines Adaptive Kernel Fusion with the proposed Progressive Receptive Fusion module, which uses temporary channel expansion, complementary depthwise receptive fields, and progressive cross-branch information transfer. We evaluate LightMIS-T, LightMIS-S, and LightMIS using five-fold cross-validation under a common nnU-Net v2.3.1 protocol on DRIVE, Kvasir-SEG, DSB18, BUSI, ISIC-2017, and ISIC-2018. Full LightMIS contains 0.131 M parameters and requires 0.575 GFLOPs for a $3\times256\times256$ input, achieving modality-macro Dice and IoU scores of 86.71% and 78.99%, respectively. Mobile U-ViT obtains 86.75% Dice and 79.07% IoU, so the observed differences are 0.04 and 0.08 percentage points. Relative to Mobile U-ViT, nnWNet, and nnU-Net, LightMIS reduces parameter count by 90.58$-$99.61% and GFLOPs by 82.54$-$96.14%. On an Arm Mali-G52 MC2 GPU, all LightMIS variants achieve full GPU delegation, with median delegated latency ranging from 53.31 ms for LightMIS-T to 138.31 ms for LightMIS. These results demonstrate a favorable accuracy$-$complexity trade-off and on-device execution feasibility for the evaluated tasks. The code is publicly available at https://github.com/AndreiiArhire/LightMIS.

cs.CV↗

arXiv2026

A Shared-Backbone Approach for Multi-Task MedMNIST Classification

Multi-task biomedical classification requires models to generalize across disparate modalities and class distributions. We study 11 heterogeneous MedMNIST datasets using the harmonic mean of per-task macro-F1. We evaluate three backbones with task-specific linear heads. We identify a resolution domain shift between the MedMNIST API and evaluation environment. Resolving this inconsistency and optimizing architecture-specific regularization substantially improved performance. Our best configuration, a ConvNeXt-Tiny backbone with label smoothing, achieved a leaderboard harmonic-mean macro-F1 of 0.73294 in the Tensor Reloaded: Multi-Task MedMNIST competition, ranking sixth at the close of the official competition phase. Our implementation is publicly available at: https://github.com/GavrilStefan-Dorian/A-Shared-Backbone-Approach-for-Multi-Task-MedMNIST-Classification

cs.CV↗

arXiv2025

UAIC_Twin_Width: An Exact yet Efficient Twin-Width Algorithm

Twin-width is a recently formulated graph and matrix invariant that intuitively quantifies how far a graph is from having the structural simplicity of a co-graph. Since its introduction in 2020, twin-width has received increasing attention and has driven research leading to notable advances in algorithmic fields, including graph theory and combinatorics. The 2023 edition of the Parameterized Algorithms and Computational Experiments (PACE) Challenge aimed to fulfill the need for a diverse and consistent public benchmark encompassing various graph structures, while also collecting state-of-the-art heuristic and exact approaches to the problem. In this paper, we propose two algorithms for efficiently computing the twin-width of graphs with arbitrary structures, comprising one exact and one heuristic approach. The proposed solutions performed strongly in the competition, with the exact algorithm achieving the best student result and ranking fourth overall. We release our source code publicly to enable practical applications of our work and support further research.

cs.DS↗

arXiv2025

Learned Lightweight Smartphone ISP with Unpaired Data

The Image Signal Processor (ISP) is a fundamental component in modern smartphone cameras responsible for conversion of RAW sensor image data to RGB images with a strong focus on perceptual quality. Recent work highlights the potential of deep learning approaches and their ability to capture details with a quality increasingly close to that of professional cameras. A difficult and costly step when developing a learned ISP is the acquisition of pixel-wise aligned paired data that maps the raw captured by a smartphone camera sensor to high-quality reference images. In this work, we address this challenge by proposing a novel training method for a learnable ISP that eliminates the need for direct correspondences between raw images and ground-truth data with matching content. Our unpaired approach employs a multi-term loss function guided by adversarial training with multiple discriminators processing feature maps from pre-trained networks to maintain content structure while learning color and texture characteristics from the target RGB dataset. Using lightweight neural network architectures suitable for mobile devices as backbones, we evaluated our method on the Zurich RAW to RGB and Fujifilm UltraISP datasets. Compared to paired training methods, our unpaired learning strategy shows strong potential and achieves high fidelity across multiple evaluation metrics. The code and pre-trained models are available at https://github.com/AndreiiArhire/Learned-Lightweight-Smartphone-ISP-with-Unpaired-Data .

cs.CV↗

arXiv2023

PACE Solver Description: A Heuristic Directed Feedback Vertex Set Problem Algorithm

A feedback vertex set of a graph is a set of nodes with the property that every cycle contains at least one vertex from the set i.e. the removal of all vertices from a feedback vertex set leads to an acyclic graph. In this short paper, we describe the algorithm for finding a minimum directed feedback vertex set used by the _UAIC_ANDREIARHIRE_ solver, submitted to the heuristic track of the 2022 PACE challenge.

cs.DM↗