The morphologies of astronomical sources are highly complex, making it essential not only to classify the identified sources into their predefined categories but also to determine the sources that are most similar to a given query source. Image-based retrieval is essential, as it allows an astronomer with a source under study to ask a computer to sift through the large archived database of sources to find the most similar ones. This is of particular interest if the source under study does not fall into a "known" category (anomalous). Our work uses the trainable COSFIRE (Combination of Shifted Filter Responses) approach for image retrieval. COSFIRE filters are automatically configured to extract the hyperlocal geometric arrangements that uniquely describe the morphological characteristics of patterns of interest in a given image; in this case astronomical sources. This is achieved by automatically examining the shape properties of a given prototype source in an image, which ultimately determines the selectivity of a COSFIRE filter. We further utilize hashing techniques, which are efficient in terms of required computation and storage, enabling scalability in handling large data sets in the image retrieval process. We evaluated the effectiveness of our approach by conducting experiments on a benchmark data set of radio galaxies, containing 1,180 training images and 404 test images. Notably, our approach achieved a mean average precision of 91% for image retrieval, surpassing the performance of the competing DenseNet-based method. Moreover, the COSFIRE filters are significantly more computationally efficient, requiring $\sim\!14\times$ fewer operations than the DenseNet-based method.
Visual localization involves estimating a query image's 6-DoF (degrees of freedom) camera pose, which is a fundamental component in various computer vision and robotic tasks. This paper presents LoGS, a vision-based localization pipeline utilizing the 3D Gaussian Splatting (GS) technique as scene representation. This novel representation allows high-quality novel view synthesis. During the mapping phase, structure-from-motion (SfM) is applied first, followed by the generation of a GS map. During localization, the initial position is obtained through image retrieval, local feature matching coupled with a PnP solver, and then a high-precision pose is achieved through the analysis-by-synthesis manner on the GS map. Experimental results on four large-scale datasets demonstrate the proposed approach's SoTA accuracy in estimating camera poses and robustness under challenging few-shot conditions.
A text encoder within Vision-Language Models (VLMs) like CLIP plays a crucial role in translating textual input into an embedding space shared with images, thereby facilitating the interpretative analysis of vision tasks through natural language. Despite the varying significance of different textual elements within a sentence depending on the context, efforts to account for variation of importance in constructing text embeddings have been lacking. We propose a framework of Semantic Token Reweighting to build Interpretable text embeddings (SToRI), which incorporates controllability as well. SToRI refines the text encoding process in CLIP by differentially weighting semantic elements based on contextual importance, enabling finer control over emphasis responsive to data-driven insights and user preferences. The efficacy of SToRI is demonstrated through comprehensive experiments on few-shot image classification and image retrieval tailored to user preferences.
Supervised Person Re-identification (Person ReID) methods have achieved excellent performance when training and testing within one camera network. However, they usually suffer from considerable performance degradation when applied to different camera systems. In recent years, many Domain Adaptation Person ReID methods have been proposed, achieving impressive performance without requiring labeled data from the target domain. However, these approaches still need the unlabeled data of the target domain during the training process, making them impractical in many real-world scenarios. Our work focuses on the more practical Domain Generalized Person Re-identification (DG-ReID) problem. Given one or more source domains, it aims to learn a generalized model that can be applied to unseen target domains. One promising research direction in DG-ReID is the use of implicit deep semantic feature expansion, and our previous method, Domain Embedding Expansion (DEX), is one such example that achieves powerful results in DG-ReID. However, in this work we show that DEX and other similar implicit deep semantic feature expansion methods, due to limitations in their proposed loss function, fail to reach their full potential on large evaluation benchmarks as they have a tendency to saturate too early. Leveraging on this analysis, we propose Unified Deep Semantic Expansion, our novel framework that unifies implicit and explicit semantic feature expansion techniques in a single framework to mitigate this early over-fitting and achieve a new state-of-the-art (SOTA) in all DG-ReID benchmarks. Further, we apply our method on more general image retrieval tasks, also surpassing the current SOTA in all of these benchmarks by wide margins.
Recent advancements in Vision-Language Models (VLMs) have enabled complex multimodal tasks by processing text and image data simultaneously, significantly enhancing the field of artificial intelligence. However, these models often exhibit biases that can skew outputs towards societal stereotypes, thus necessitating debiasing strategies. Existing debiasing methods focus narrowly on specific modalities or tasks, and require extensive retraining. To address these limitations, this paper introduces Selective Feature Imputation for Debiasing (SFID), a novel methodology that integrates feature pruning and low confidence imputation (LCI) to effectively reduce biases in VLMs. SFID is versatile, maintaining the semantic integrity of outputs and costly effective by eliminating the need for retraining. Our experimental results demonstrate SFID's effectiveness across various VLMs tasks including zero-shot classification, text-to-image retrieval, image captioning, and text-to-image generation, by significantly reducing gender biases without compromising performance. This approach not only enhances the fairness of VLMs applications but also preserves their efficiency and utility across diverse scenarios.
Image retrieval is crucial in robotics and computer vision, with downstream applications in robot place recognition and vision-based product recommendations. Modern retrieval systems face two key challenges: scalability and efficiency. State-of-the-art image retrieval systems train specific neural networks for each dataset, an approach that lacks scalability. Furthermore, since retrieval speed is directly proportional to embedding size, existing systems that use large embeddings lack efficiency. To tackle scalability, recent works propose using off-the-shelf foundation models. However, these models, though applicable across datasets, fall short in achieving performance comparable to that of dataset-specific models. Our key observation is that, while foundation models capture necessary subtleties for effective retrieval, the underlying distribution of their embedding space can negatively impact cosine similarity searches. We introduce Autoencoders with Strong Variance Constraints (AE-SVC), which, when used for projection, significantly improves the performance of foundation models. We provide an in-depth theoretical analysis of AE-SVC. Addressing efficiency, we introduce Single-shot Similarity Space Distillation ((SS)$_2$D), a novel approach to learn embeddings with adaptive sizes that offers a better trade-off between size and performance. We conducted extensive experiments on four retrieval datasets, including Stanford Online Products (SoP) and Pittsburgh30k, using four different off-the-shelf foundation models, including DinoV2 and CLIP. AE-SVC demonstrates up to a $16\%$ improvement in retrieval performance, while (SS)$_2$D shows a further $10\%$ improvement for smaller embedding sizes.
Content-Based Image Retrieval (CBIR) have shown promising results in the field of medical diagnosis, which aims to provide support to medical professionals (doctor or pathologist). However, the ultimate decision regarding the diagnosis is made by the medical professional, drawing upon their accumulated experience. In this context, we believe that artificial intelligence can play a pivotal role in addressing the challenges in medical diagnosis not by making the final decision but by assisting in the diagnosis process with the most relevant information. The CBIR methods use similarity metrics to compare feature vectors generated from images using Convolutional Neural Networks (CNNs). In addition to the information contained in medical images, clinical data about the patient is often available and is also relevant in the final decision-making process by medical professionals. In this paper, we propose a novel method named CBIDR, which leverage both medical images and clinical data of patient, combining them through the ranking algorithm TOPSIS. The goal is to aid medical professionals in their final diagnosis by retrieving images and clinical data of patient that are most similar to query data from the database. As a case study, we illustrate our CBIDR for diagnostic of oral cancer including histopathological images and clinical data of patient. Experimental results in terms of accuracy achieved 97.44% in Top-1 and 100% in Top-5 showing the effectiveness of the proposed approach.
Vision-Language Models (VLMs) have transformed tasks requiring visual and reasoning abilities, such as image retrieval and Visual Question Answering (VQA). Despite their success, VLMs face significant challenges with tasks involving geometric reasoning, algebraic problem-solving, and counting. These limitations stem from difficulties effectively integrating multiple modalities and accurately interpreting geometry-related tasks. Various works claim that introducing a captioning pipeline before VQA tasks enhances performance. We incorporated this pipeline for tasks involving geometry, algebra, and counting. We found that captioning results are not generalizable, specifically with larger VLMs primarily trained on downstream QnA tasks showing random performance on math-related challenges. However, we present a promising alternative: task-based prompting, enriching the prompt with task-specific guidance. This approach shows promise and proves more effective than direct captioning methods for math-heavy problems.
Multimodal models, which combine visual and textual information, have recently gained significant recognition. This paper addresses the multimodal challenge of Text-Image retrieval and introduces a novel task that extends the modalities to include temporal data. The Temporal Image Caption Retrieval Competition (TICRC) presented in this paper is based on the Chronicling America and Challenging America projects, which offer access to an extensive collection of digitized historic American newspapers spanning 274 years. In addition to the competition results, we provide an analysis of the delivered dataset and the process of its creation.
In this work, we present MedImageInsight, an open-source medical imaging embedding model. MedImageInsight is trained on medical images with associated text and labels across a diverse collection of domains, including X-Ray, CT, MRI, dermoscopy, OCT, fundus photography, ultrasound, histopathology, and mammography. Rigorous evaluations demonstrate MedImageInsight's ability to achieve state-of-the-art (SOTA) or human expert level performance across classification, image-image search, and fine-tuning tasks. Specifically, on public datasets, MedImageInsight achieves SOTA in CT 3D medical image retrieval, as well as SOTA in disease classification and search for chest X-ray, dermatology, and OCT imaging. Furthermore, MedImageInsight achieves human expert performance in bone age estimation (on both public and partner data), as well as AUC above 0.9 in most other domains. When paired with a text decoder, MedImageInsight achieves near SOTA level single image report findings generation with less than 10\% the parameters of other models. Compared to fine-tuning GPT-4o with only MIMIC-CXR data for the same task, MedImageInsight outperforms in clinical metrics, but underperforms on lexical metrics where GPT-4o sets a new SOTA. Importantly for regulatory purposes, MedImageInsight can generate ROC curves, adjust sensitivity and specificity based on clinical need, and provide evidence-based decision support through image-image search (which can also enable retrieval augmented generation). In an independent clinical evaluation of image-image search in chest X-ray, MedImageInsight outperformed every other publicly available foundation model evaluated by large margins (over 6 points AUC), and significantly outperformed other models in terms of AI fairness (across age and gender). We hope releasing MedImageInsight will help enhance collective progress in medical imaging AI research and development.
Understanding long text is of great demands in practice but beyond the reach of most language-image pre-training (LIP) models. In this work, we empirically confirm that the key reason causing such an issue is that the training images are usually paired with short captions, leaving certain tokens easily overshadowed by salient tokens. Towards this problem, our initial attempt is to relabel the data with long captions, however, directly learning with which may lead to performance degradation in understanding short text (e.g., in the image classification task). Then, after incorporating corner tokens to aggregate diverse textual information, we manage to help the model catch up to its original level of short text understanding yet greatly enhance its capability of long text understanding. We further look into whether the model can continuously benefit from longer captions and notice a clear trade-off between the performance and the efficiency. Finally, we validate the effectiveness of our approach using a self-constructed large-scale dataset, which consists of 100M long caption oriented text-image pairs. Our method demonstrates superior performance in long-text-image retrieval tasks. The project page is available at https://wuw2019.github.io/lot-lip.
Despite the prevalence and historical importance of maps in digital collections, current methods of navigating and exploring map collections are largely restricted to catalog records and structured metadata. In this paper, we explore the potential for interactively searching large-scale map collections using natural language inputs ("maps with sea monsters"), visual inputs (i.e., reverse image search), and multimodal inputs (an example map + "more grayscale"). As a case study, we adopt 562,842 images of maps publicly accessible via the Library of Congress's API. To accomplish this, we use the mulitmodal Contrastive Language-Image Pre-training (CLIP) machine learning model to generate embeddings for these maps, and we develop code to implement exploratory search capabilities with these input strategies. We present results for example searches created in consultation with staff in the Library of Congress's Geography and Map Division and describe the strengths, weaknesses, and possibilities for these search queries. Moreover, we introduce a fine-tuning dataset of 10,504 map-caption pairs, along with an architecture for fine-tuning a CLIP model on this dataset. To facilitate re-use, we provide all of our code in documented, interactive Jupyter notebooks and place all code into the public domain. Lastly, we discuss the opportunities and challenges for applying these approaches across both digitized and born-digital collections held by galleries, libraries, archives, and museums.
Scene sketch semantic segmentation is a crucial task for various applications including sketch-to-image retrieval and scene understanding. Existing sketch segmentation methods treat sketches as bitmap images, leading to the loss of temporal order among strokes due to the shift from vector to image format. Moreover, these methods struggle to segment objects from categories absent in the training data. In this paper, we propose a Class-Agnostic Visio-Temporal Network (CAVT) for scene sketch semantic segmentation. CAVT employs a class-agnostic object detector to detect individual objects in a scene and groups the strokes of instances through its post-processing module. This is the first approach that performs segmentation at both the instance and stroke levels within scene sketches. Furthermore, there is a lack of free-hand scene sketch datasets with both instance and stroke-level class annotations. To fill this gap, we collected the largest Free-hand Instance- and Stroke-level Scene Sketch Dataset (FrISS) that contains 1K scene sketches and covers 403 object classes with dense annotations. Extensive experiments on FrISS and other datasets demonstrate the superior performance of our method over state-of-the-art scene sketch segmentation models. The code and dataset will be made public after acceptance.
The intersection of Artificial Intelligence and Digital Humanities enables researchers to explore cultural heritage collections with greater depth and scale. In this paper, we present EUFCC-CIR, a dataset designed for Composed Image Retrieval (CIR) within Galleries, Libraries, Archives, and Museums (GLAM) collections. Our dataset is built on top of the EUFCC-340K image labeling dataset and contains over 180K annotated CIR triplets. Each triplet is composed of a multi-modal query (an input image plus a short text describing the desired attribute manipulations) and a set of relevant target images. The EUFCC-CIR dataset fills an existing gap in CIR-specific resources for Digital Humanities. We demonstrate the value of the EUFCC-CIR dataset by highlighting its unique qualities in comparison to other existing CIR datasets and evaluating the performance of several zero-shot CIR baselines.
Image similarity metrics play an important role in computer vision applications, as they are used in image processing, computer vision and machine learning. Furthermore, those metrics enable tasks such as image retrieval, object recognition and quality assessment, essential in fields like healthcare, astronomy and surveillance. Existing metrics, such as PSNR, MSE, SSIM, ISSM and FSIM, often face limitations in terms of either speed, complexity or sensitivity to small changes in images. To address these challenges, a novel image similarity metric, namely CSIM, that combines real-time while being sensitive to subtle image variations is investigated in this paper. The novel metric uses Gaussian Copula from probability theory to transform an image into vectors of pixel distribution associated to local image patches. These vectors contain, in addition to intensities and pixel positions, information on the dependencies between pixel values, capturing the structural relationships within the image. By leveraging the properties of Copulas, CSIM effectively models the joint distribution of pixel intensities, enabling a more nuanced comparison of image patches making it more sensitive to local changes compared to other metrics. Experimental results demonstrate that CSIM outperforms existing similarity metrics in various image distortion scenarios, including noise, compression artifacts and blur. The metric's ability to detect subtle differences makes it suitable for applications requiring high precision, such as medical imaging, where the detection of minor anomalies can be of a high importance. The results obtained in this work can be reproduced from this Github repository: https://github.com/safouaneelg/copulasimilarity.
Visual Place Recognition (VPR) is a crucial component of many visual localization pipelines for embodied agents. VPR is often formulated as an image retrieval task aimed at jointly learning local features and an aggregation method. The current state-of-the-art VPR methods rely on VLAD aggregation, which can be trained to learn a weighted contribution of features through their soft assignment to cluster centers. However, this process has two key limitations. Firstly, the feature-to-cluster weighting does not account for over-represented repetitive structures within a cluster, e.g., shadows or window panes; this phenomenon is also referred to as the `burstiness' problem, classically solved by discounting repetitive features before aggregation. Secondly, feature to cluster comparisons are compute-intensive for state-of-the-art image encoders with high-dimensional local features. This paper addresses these limitations by introducing VLAD-BuFF with two novel contributions: i) a self-similarity based feature discounting mechanism to learn Burst-aware features within end-to-end VPR training, and ii) Fast Feature aggregation by reducing local feature dimensions specifically through PCA-initialized learnable pre-projection. We benchmark our method on 9 public datasets, where VLAD-BuFF sets a new state of the art. Our method is able to maintain its high recall even for 12x reduced local feature dimensions, thus enabling fast feature aggregation without compromising on recall. Through additional qualitative studies, we show how our proposed weighting method effectively downweights the non-distinctive features. Source code: https://github.com/Ahmedest61/VLAD-BuFF/.
In this paper, we introduce SearchDet, a training-free long-tail object detection framework that significantly enhances open-vocabulary object detection performance. SearchDet retrieves a set of positive and negative images of an object to ground, embeds these images, and computes an input image-weighted query which is used to detect the desired concept in the image. Our proposed method is simple and training-free, yet achieves over 48.7% mAP improvement on ODinW and 59.1% mAP improvement on LVIS compared to state-of-the-art models such as GroundingDINO. We further show that our approach of basing object detection on a set of Web-retrieved exemplars is stable with respect to variations in the exemplars, suggesting a path towards eliminating costly data annotation and training procedures.
Current image retrieval systems often face domain specificity and generalization issues. This study aims to overcome these limitations by developing a computationally efficient training framework for a universal feature extractor that provides strong semantic image representations across various domains. To this end, we curated a multi-domain training dataset, called M4D-35k, which allows for resource-efficient training. Additionally, we conduct an extensive evaluation and comparison of various state-of-the-art visual-semantic foundation models and margin-based metric learning loss functions regarding their suitability for efficient universal feature extraction. Despite constrained computational resources, we achieve near state-of-the-art results on the Google Universal Image Embedding Challenge, with a mMP@5 of 0.721. This places our method at the second rank on the leaderboard, just 0.7 percentage points behind the best performing method. However, our model has 32% fewer overall parameters and 289 times fewer trainable parameters. Compared to methods with similar computational requirements, we outperform the previous state of the art by 3.3 percentage points. We release our code and M4D-35k training set annotations at https://github.com/morrisfl/UniFEx.
Cross-View Geo-Localisation within urban regions is challenging in part due to the lack of geo-spatial structuring within current datasets and techniques. We propose utilising graph representations to model sequences of local observations and the connectivity of the target location. Modelling as a graph enables generating previously unseen sequences by sampling with new parameter configurations. To leverage this newly available information, we propose a GNN-based architecture, producing spatially strong embeddings and improving discriminability over isolated image embeddings. We outline SpaGBOL, introducing three novel contributions. 1) The first graph-structured dataset for Cross-View Geo-Localisation, containing multiple streetview images per node to improve generalisation. 2) Introducing GNNs to the problem, we develop the first system that exploits the correlation between node proximity and feature similarity. 3) Leveraging the unique properties of the graph representation - we demonstrate a novel retrieval filtering approach based on neighbourhood bearings. SpaGBOL achieves state-of-the-art accuracies on the unseen test graph - with relative Top-1 retrieval improvements on previous techniques of 11%, and 50% when filtering with Bearing Vector Matching on the SpaGBOL dataset.
Despite the impressive performance of autoregressive Language Models (LM) it has been shown that due to reporting bias, LMs lack visual knowledge, i.e. they do not know much about the visual world and its properties. To augment LMs with visual knowledge, existing solutions often rely on explicit images, requiring time-consuming retrieval or image generation systems. This paper shows that explicit images are not necessary to visually augment an LM. Instead, we use visually-grounded text representations obtained from the well-known CLIP multimodal system. For a fair comparison, we modify VALM, a visually-augmented LM which uses image retrieval and representation, to work directly with visually-grounded text representations. We name this new model BLIND-VALM. We show that BLIND-VALM performs on par with VALM for Visual Language Understanding (VLU), Natural Language Understanding (NLU) and Language Modeling tasks, despite being significantly more efficient and simpler. We also show that scaling up our model within the compute budget of VALM, either increasing the model or pre-training corpus size, we outperform VALM for all the evaluation tasks.