arXiv subjects

Rozenn Dahyot

Publications and source records attributed to Rozenn Dahyot.

At least 19 recordsLinked to original sources

arXiv2025

Benchmarking SAM2-based Trackers on FMOX

Several object tracking pipelines extending Segment Anything Model 2 (SAM2) have been proposed in the past year, where the approach is to follow and segment the object from a single exemplar template provided by the user on a initialization frame. We propose to benchmark these high performing trackers (SAM2, EfficientTAM, DAM4SAM and SAMURAI) on datasets containing fast moving objects (FMO) specifically designed to be challenging for tracking approaches. The goal is to understand better current limitations in state-of-the-art trackers by providing more detailed insights on the behavior of these trackers. We show that overall the trackers DAM4SAM and SAMURAI perform well on more challenging sequences.

cs.CV↗

arXiv2025

Benchmarking EfficientTAM on FMO datasets

Fast and tiny object tracking remains a challenge in computer vision and in this paper we first introduce a JSON metadata file associated with four open source datasets of Fast Moving Objects (FMOs) image sequences. In addition, we extend the description of the FMOs datasets with additional ground truth information in JSON format (called FMOX) with object size information. Finally we use our FMOX file to test a recently proposed foundational model for tracking (called EfficientTAM) showing that its performance compares well with the pipelines originally taylored for these FMO datasets. Our comparison of these state-of-the-art techniques on FMOX is provided with Trajectory Intersection of Union (TIoU) scores. The code and JSON is shared open source allowing FMOX to be accessible and usable for other machine learning pipelines aiming to process FMO datasets.

cs.CV↗

arXiv2024

Performance of Gaussian Mixture Model Classifiers on Embedded Feature Spaces

Data embeddings with CLIP and ImageBind provide powerful features for the analysis of multimedia and/or multimodal data. We assess their performance here for classification using a Gaussian Mixture models (GMMs) based layer as an alternative to the standard Softmax layer. GMMs based classifiers have recently been shown to have interesting performances as part of deep learning pipelines trained end-to-end. Our first contribution is to investigate GMM based classification performance taking advantage of the embedded spaces CLIP and ImageBind. Our second contribution is in proposing our own GMM based classifier with a lower parameters count than previously proposed. Our findings are, that in most cases, on these tested embedded spaces, one gaussian component in the GMMs is often enough for capturing each class, and we hypothesize that this may be due to the contrastive loss used for training these embedded spaces that naturally concentrates features together for each class. We also observed that ImageBind often provides better performance than CLIP for classification of image datasets even when these embedded spaces are compressed using PCA.

cs.CV↗

arXiv2023

Model-based inexact graph matching on top of CNNs for semantic scene understanding

Deep learning based pipelines for semantic segmentation often ignore structural information available on annotated images used for training. We propose a novel post-processing module enforcing structural knowledge about the objects of interest to improve segmentation results provided by deep learning. This module corresponds to a "many-to-one-or-none" inexact graph matching approach, and is formulated as a quadratic assignment problem. Our approach is compared to a CNN-based segmentation (for various CNN backbones) on two public datasets, one for face segmentation from 2D RGB images (FASSEG), and the other for brain segmentation from 3D MRIs (IBSR). Evaluations are performed using two types of structural information (distances and directional relations, , this choice being a hyper-parameter of our generic framework). On FASSEG data, results show that our module improves accuracy of the CNN by about 6.3% (the Hausdorff distance decreases from 22.11 to 20.71). On IBSR data, the improvement is of 51% (the Hausdorff distance decreases from 11.01 to 5.4). In addition, our approach is shown to be resilient to small training datasets that often limit the performance of deep learning methods: the improvement increases as the size of the training dataset decreases.

cs.CV↗

arXiv2023

Combining geolocation and height estimation of objects from street level imagery

We propose a pipeline for combined multi-class object geolocation and height estimation from street level RGB imagery, which is considered as a single available input data modality. Our solution is formulated via Markov Random Field optimization with deterministic output. The proposed technique uses image metadata along with coordinates of objects detected in the image plane as found by a custom-trained Convolutional Neural Network. Computing the object height using our methodology, in addition to object geolocation, has negligible effect on the overall computational cost. Accuracy is demonstrated experimentally for water drains and road signs on which we achieve average elevation estimation error lower than 20cm.

cs.CV↗

arXiv2022

Principal Component Classification

We propose to directly compute classification estimates by learning features encoded with their class scores using PCA. Our resulting model has a encoder-decoder structure suitable for supervised learning, it is computationally efficient and performs well for classification on several datasets.

cs.LG↗

arXiv2022

Harmonic Convolutional Networks based on Discrete Cosine Transform

Convolutional neural networks (CNNs) learn filters in order to capture local correlation patterns in feature space. We propose to learn these filters as combinations of preset spectral filters defined by the Discrete Cosine Transform (DCT). Our proposed DCT-based harmonic blocks replace conventional convolutional layers to produce partially or fully harmonic versions of new or existing CNN architectures. Using DCT energy compaction properties, we demonstrate how the harmonic networks can be efficiently compressed by truncating high-frequency information in harmonic blocks thanks to the redundancies in the spectral domain. We report extensive experimental validation demonstrating benefits of the introduction of harmonic blocks into state-of-the-art CNN models in image classification, object detection and semantic segmentation applications.

cs.CV↗

arXiv2022

DR-VNet: Retinal Vessel Segmentation via Dense Residual UNet

Accurate retinal vessel segmentation is an important task for many computer-aided diagnosis systems. Yet, it is still a challenging problem due to the complex vessel structures of an eye. Numerous vessel segmentation methods have been proposed recently, however more research is needed to deal with poor segmentation of thin and tiny vessels. To address this, we propose a new deep learning pipeline combining the efficiency of residual dense net blocks and, residual squeeze and excitation blocks. We validate experimentally our approach on three datasets and show that our pipeline outperforms current state of the art techniques on the sensitivity metric relevant to assess capture of small vessels.

eess.IV↗

arXiv2021

Context Aware Object Geotagging

Localization of street objects from images has gained a lot of attention in recent years. We propose an approach to improve asset geolocation from street view imagery by enhancing the quality of the metadata associated with the images using Structure from Motion. The predicted object geolocation is further refined by imposing contextual geographic information extracted from OpenStreetMap. Our pipeline is validated experimentally against the state of the art approaches for geotagging traffic lights.

cs.LG↗

arXiv2021

3D point cloud segmentation using GIS

In this paper we propose an approach to perform semantic segmentation of 3D point cloud data by importing the geographic information from a 2D GIS layer (OpenStreetMap). The proposed automatic procedure identifies meaningful units such as buildings and adjusts their locations to achieve best fit between the GIS polygonal perimeters and the point cloud. Our processing pipeline is presented and illustrated by segmenting point cloud data of Trinity College Dublin (Ireland) campus constructed from optical imagery collected by a drone.

cs.CV↗

arXiv2021

Sliced $\mathcal{L}_2$ Distance for Colour Grading

We propose a new method with $\mathcal{L}_2$ distance that maps one $N$-dimensional distribution to another, taking into account available information about correspondences. We solve the high-dimensional problem in 1D space using an iterative projection approach. To show the potentials of this mapping, we apply it to colour transfer between two images that exhibit overlapped scenes. Experiments show quantitative and qualitative competitive results as compared with the state of the art colour transfer methods.

cs.CV↗

arXiv2020

Tensor Reordering for CNN Compression

We show how parameter redundancy in Convolutional Neural Network (CNN) filters can be effectively reduced by pruning in spectral domain. Specifically, the representation extracted via Discrete Cosine Transform (DCT) is more conducive for pruning than the original space. By relying on a combination of weight tensor reshaping and reordering we achieve high levels of layer compression with just minor accuracy loss. Our approach is applied to compress pretrained CNNs and we show that minor additional fine-tuning allows our method to recover the original model performance after a significant parameter reduction. We validate our approach on ResNet-50 and MobileNet-V2 architectures for ImageNet classification task.

cs.LG↗

arXiv2020

Iterative Nadaraya-Watson Distribution Transfer for Colour Grading

We propose a new method with Nadaraya-Watson that maps one N-dimensional distribution to another taking into account available information about correspondences. We extend the 2D/3D problem to higher dimensions by encoding overlapping neighborhoods of data points and solve the high dimensional problem in 1D space using an iterative projection approach. To show potentials of this mapping, we apply it to colour transfer between two images that exhibit overlapped scene. Experiments show quantitative and qualitative improvements over previous state of the art colour transfer methods.

cs.CV↗

arXiv2020

Patch based Colour Transfer using SIFT Flow

We propose a new colour transfer method with Optimal Transport (OT) to transfer the colour of a sourceimage to match the colour of a target image of the same scene that may exhibit large motion changes betweenimages. By definition OT does not take into account any available information about correspondences whencomputing the optimal solution. To tackle this problem we propose to encode overlapping neighborhoodsof pixels using both their colour and spatial correspondences estimated using motion estimation. We solvethe high dimensional problem in 1D space using an iterative projection approach. We further introducesmoothing as part of the iterative algorithms for solving optimal transport namely Iterative DistributionTransport (IDT) and its variant the Sliced Wasserstein Distance (SWD). Experiments show quantitative andqualitative improvements over previous state of the art colour transfer methods.

cs.CV↗

arXiv2020

Performance-Oriented Neural Architecture Search

Hardware-Software Co-Design is a highly successful strategy for improving performance of domain-specific computing systems. We argue for the application of the same methodology to deep learning; specifically, we propose to extend neural architecture search with information about the hardware to ensure that the model designs produced are highly efficient in addition to the typical criteria around accuracy. Using the task of keyword spotting in audio on edge computing devices, we demonstrate that our approach results in neural architecture that is not only highly accurate, but also efficiently mapped to the computing platform which will perform the inference. Using our modified neural architecture search, we demonstrate $0.88\%$ increase in TOP-1 accuracy with $1.85\times$ reduction in latency for keyword spotting in audio on an embedded SoC, and $1.59\times$ on a high-end GPU.

cs.LG↗

arXiv2019

Entropic Regularisation of Robust Optimal Transport

Grogan et al [11,12] have recently proposed a solution to colour transfer by minimising the Euclidean distance L2 between two probability density functions capturing the colour distributions of two images (palette and target). It was shown to be very competitive to alternative solutions based on Optimal Transport for colour transfer. We show that in fact Grogan et al's formulation can also be understood as a new robust Optimal Transport based framework with entropy regularisation over marginals.

cs.CV↗

arXiv2019

Harmonic Networks with Limited Training Samples

Convolutional neural networks (CNNs) are very popular nowadays for image processing. CNNs allow one to learn optimal filters in a (mostly) supervised machine learning context. However this typically requires abundant labelled training data to estimate the filter parameters. Alternative strategies have been deployed for reducing the number of parameters and / or filters to be learned and thus decrease overfitting. In the context of reverting to preset filters, we propose here a computationally efficient harmonic block that uses Discrete Cosine Transform (DCT) filters in CNNs. In this work we examine the performance of harmonic networks in limited training data scenario. We validate experimentally that its performance compares well against scattering networks that use wavelets as preset filters.

cs.CV↗

arXiv2019

Automatic detection of passable roads after floods in remote sensed and social media data

This paper addresses the problem of floods classification and floods aftermath detection utilizing both social media and satellite imagery. Automatic detection of disasters such as floods is still a very challenging task. The focus lies on identifying passable routes or roads during floods. Two novel solutions are presented, which were developed for two corresponding tasks at the MediaEval 2018 benchmarking challenge. The tasks are (i) identification of images providing evidence for road passability and (ii) differentiation and detection of passable and non-passable roads in images from two complementary sources of information. For the first challenge, we mainly rely on object and scene-level features extracted through multiple deep models pre-trained on the ImageNet and Places datasets. The object and scene-level features are then combined using early, late and double fusion techniques. To identify whether or not it is possible for a vehicle to pass a road in satellite images, we rely on Convolutional Neural Networks and a transfer learning-based classification approach. The evaluation of the proposed methods are carried out on the large-scale datasets provided for the benchmark competition. The results demonstrate significant improvement in the performance over the recent state-of-art approaches.

cs.IR↗