Building models that comprehends videos and responds specific user instructions is a practical and challenging topic, as it requires mastery of both vision understanding and knowledge reasoning. Compared to language and image modalities, training efficiency remains a serious problem as existing studies train models on massive sparse videos paired with brief descriptions. In this paper, we introduce \textbf{VidCoM}, a fast adaptive framework that leverages Large Language Models (LLMs) to reason about videos using lightweight visual tools. Specifically, we reveal that the key to responding to specific instructions is focusing on relevant video events, and utilize two visual tools, structured scene graph generation and descriptive image caption generation, to gather and represent the event information. Thus, a LLM enriched with world knowledge is adopted as the reasoning agent to achieve the responses by performing multiple reasoning steps on specific video events. To address the difficulty of LLMs identifying video events, we further propose an Instruction-oriented Video Events Recognition (InsOVER) algorithm. This algorithm locates the corresponding video events based on an efficient Hungarian matching between decompositions of linguistic instructions and video events, thereby enabling LLMs to interact effectively with extended videos. Extensive experiments on two typical video comprehension tasks show that the proposed tuning-free framework outperforms the pre-trained models including Flamingo-80B, to achieve the state-of-the-art performance. Our source code and system will be publicly available.
Out-of-distribution (OOD) detection is essential for reliable and trustworthy machine learning. Recent multi-modal OOD detection leverages textual information from in-distribution (ID) class names for visual OOD detection, yet it currently neglects the rich contextual information of ID classes. Large language models (LLMs) encode a wealth of world knowledge and can be prompted to generate descriptive features for each class. Indiscriminately using such knowledge causes catastrophic damage to OOD detection due to LLMs' hallucinations, as is observed by our analysis. In this paper, we propose to apply world knowledge to enhance OOD detection performance through selective generation from LLMs. Specifically, we introduce a consistency-based uncertainty calibration method to estimate the confidence score of each generation. We further extract visual objects from each image to fully capitalize on the aforementioned world knowledge. Extensive experiments demonstrate that our method consistently outperforms the state-of-the-art.
Increasing cultivated lands, crop homogenization and global food trade have fostered the spread of crop pests and diseases. Optimizing crop protection is urgently needed to ensure food safety. One aspect of crop protection is surveillance, which focuses on the early detection of a parasite, and control, aiming to fight and possibly eradicate it. Network theory has been widely used to model the spread of human and animal infectious diseases in systems described through nodes and edges. It has been successfully used to optimize monitoring and immunization campaigns. In crop protection, there is a growing literature using this theory to describe parasites spread and to conceive protection strategies. Here we review the use of network theory in crop protection, from the more descriptive to the more applied approaches aimed to optimize crop protection. We retrace the logical process that has led epidemiological models to rely on network theory, and we provide examples of how the spread of crop parasites has been represented via a network description. We define the objectives of surveillance and control, and we show how these have been declined in the network-based epidemiological sphere and then adapted in the agricultural context. We eventually discuss the discrepancy between the application of network theory in surveillance and control to identify culprits and solutions. We find that: i) scientists have successfully interpreted very different modes of parasitic transmission under the lens of network theory; ii) while network-based surveillance has progressively clarified its objectives and sound tools have been proposed, network-based control has been less studied and applied; iii) network-thinking must address how to properly define edges and nodes at different geographic scale to broad its application in crop protection.
In the field of medical Vision-Language Pre-training (VLP), significant efforts have been devoted to deriving text and image features from both clinical reports and associated medical images. However, most existing methods may have overlooked the opportunity in leveraging the inherent hierarchical structure of clinical reports, which are generally split into `findings' for descriptive content and `impressions' for conclusive observation. Instead of utilizing this rich, structured format, current medical VLP approaches often simplify the report into either a unified entity or fragmented tokens. In this work, we propose a novel clinical prior guided VLP framework named IMITATE to learn the structure information from medical reports with hierarchical vision-language alignment. The framework derives multi-level visual features from the chest X-ray (CXR) images and separately aligns these features with the descriptive and the conclusive text encoded in the hierarchical medical report. Furthermore, a new clinical-informed contrastive loss is introduced for cross-modal learning, which accounts for clinical prior knowledge in formulating sample correlations in contrastive learning. The proposed model, IMITATE, outperforms baseline VLP methods across six different datasets, spanning five medical imaging downstream tasks. Comprehensive experimental results highlight the advantages of integrating the hierarchical structure of medical reports for vision-language alignment. The code related to this paper is available at https://github.com/cheliu-computation/IMITATE-TMI2024.
Image captioning is a challenging task involving generating a textual description for an image using computer vision and natural language processing techniques. This paper proposes a deep neural framework for image caption generation using a GRU-based attention mechanism. Our approach employs multiple pre-trained convolutional neural networks as the encoder to extract features from the image and a GRU-based language model as the decoder to generate descriptive sentences. To improve performance, we integrate the Bahdanau attention model with the GRU decoder to enable learning to focus on specific image parts. We evaluate our approach using the MSCOCO and Flickr30k datasets and show that it achieves competitive scores compared to state-of-the-art methods. Our proposed framework can bridge the gap between computer vision and natural language and can be extended to specific domains.
Interpretability plays a critical role in the application of statistical learning for estimating heterogeneous treatment effects (HTE) for complex diseases. In this study, we leverage a rule-based workflow, namely causal rule learning (CRL) to estimate and enhance our understanding of HTE for atrial septal defect, addressing an overlooked question in previous literature: what if an individual simultaneously belongs to multiple groups with different average treatment effects? The CRL process consists of three steps: rule discovery, which generates a set of causal rules with corresponding subgroup average treatment effects; rule selection, which identifies a subset of these rules to deconstruct individual-level treatment effects as a linear combination of subgroup-level effects; and rule analysis, which outlines a detailed procedure for further analyzing each selected rule from multiple perspectives to identify the most promising rules for validation. Extensive simulation studies and real-world data analysis demonstrate that CRL outperforms other methods in providing interpretable estimates of HTE, especially when dealing with complex ground truth and sufficient sample sizes.
Estimating homography matrix between two images has various applications like image stitching or image mosaicing and spatial information retrieval from multiple camera views, but has been proved to be a complicated problem, especially in cases of radically different camera poses and zoom factors. Many relevant approaches have been proposed, utilizing direct feature based, or deep learning methodologies. In this paper, we propose a generalized RANSAC algorithm, H-RANSAC, to retrieve homography image transformations from sets of points without descriptive local feature vectors and point pairing. We allow the points to be optionally labelled in two classes. We propose a robust criterion that rejects implausible point selection before each iteration of RANSAC, based on the type of the quadrilaterals formed by random point pair selection (convex or concave and (non)-self-intersecting). A similar post-hoc criterion rejects implausible homography transformations is included at the end of each iteration. The expected maximum iterations of $H$-RANSAC are derived for different probabilities of success, according to the number of points per image and per class, and the percentage of outliers. The proposed methodology is tested on a large dataset of images acquired by 12 cameras during real football matches, where radically different views at each timestamp are to be matched. Comparisons with state-of-the-art implementations of RANSAC combined with classic and deep learning image salient point detection indicates the superiority of the proposed $H$-RANSAC, in terms of average reprojection error and number of successfully processed pairs of frames, rendering it the method of choice in cases of image homography alignment with few tens of points, while local features are not available, or not descriptive enough. The implementation of $H$-RANSAC is available in https://github.com/gnousias/H-RANSAC
We present Soda (Symbolic Objective Descriptive Analysis), a language that helps to treat qualities and quantities in a natural way and greatly simplifies the task of checking their correctness. We present key properties for the language motivated by the design of a descriptive language to encode complex requirements on computer systems, and we explain how these key properties must be addressed to model these requirements with simple definitions. We give an overview of a tool that helps to describe problems in an easy way that we consider more transparent and less error-prone.
Neural Image Classifiers are effective but inherently hard to interpret and susceptible to adversarial attacks. Solutions to both problems exist, among others, in the form of counterfactual examples generation to enhance explainability or adversarially augment training datasets for improved robustness. However, existing methods exclusively address only one of the issues. We propose a unified framework leveraging image-to-image translation Generative Adversarial Networks (GANs) to produce counterfactual samples that highlight salient regions for interpretability and act as adversarial samples to augment the dataset for more robustness. This is achieved by combining the classifier and discriminator into a single model that attributes real images to their respective classes and flags generated images as "fake". We assess the method's effectiveness by evaluating (i) the produced explainability masks on a semantic segmentation task for concrete cracks and (ii) the model's resilience against the Projected Gradient Descent (PGD) attack on a fruit defects detection problem. Our produced saliency maps are highly descriptive, achieving competitive IoU values compared to classical segmentation models despite being trained exclusively on classification labels. Furthermore, the model exhibits improved robustness to adversarial attacks, and we show how the discriminator's "fakeness" value serves as an uncertainty measure of the predictions.
Nuanced understanding and the generation of detailed descriptive content for (bimanual) manipulation actions in videos is important for disciplines such as robotics, human-computer interaction, and video content analysis. This study describes a novel method, integrating graph based modeling with layered hierarchical attention mechanisms, resulting in higher precision and better comprehensiveness of video descriptions. To achieve this, we encode, first, the spatio-temporal inter dependencies between objects and actions with scene graphs and we combine this, in a second step, with a novel 3-level architecture creating a hierarchical attention mechanism using Graph Attention Networks (GATs). The 3-level GAT architecture allows recognizing local, but also global contextual elements. This way several descriptions with different semantic complexity can be generated in parallel for the same video clip, enhancing the discriminative accuracy of action recognition and action description. The performance of our approach is empirically tested using several 2D and 3D datasets. By comparing our method to the state of the art we consistently obtain better performance concerning accuracy, precision, and contextual relevance when evaluating action recognition as well as description generation. In a large set of ablation experiments we also assess the role of the different components of our model. With our multi-level approach the system obtains different semantic description depths, often observed in descriptions made by different people, too. Furthermore, better insight into bimanual hand-object interactions as achieved by our model may portend advancements in the field of robotics, enabling the emulation of intricate human actions with heightened precision.
Session-based recommendation is devoted to characterizing preferences of anonymous users based on short sessions. Existing methods mostly focus on mining limited item co-occurrence patterns exposed by item ID within sessions, while ignoring what attracts users to engage with certain items is rich multi-modal information displayed on pages. Generally, the multi-modal information can be classified into two categories: descriptive information (e.g., item images and description text) and numerical information (e.g., price). In this paper, we aim to improve session-based recommendation by modeling the above multi-modal information holistically. There are mainly three issues to reveal user intent from multi-modal information: (1) How to extract relevant semantics from heterogeneous descriptive information with different noise? (2) How to fuse these heterogeneous descriptive information to comprehensively infer user interests? (3) How to handle probabilistic influence of numerical information on user behaviors? To solve above issues, we propose a novel multi-modal session-based recommendation (MMSBR) that models both descriptive and numerical information under a unified framework. Specifically, a pseudo-modality contrastive learning is devised to enhance the representation learning of descriptive information. Afterwards, a hierarchical pivot transformer is presented to fuse heterogeneous descriptive information. Moreover, we represent numerical information with Gaussian distribution and design a Wasserstein self-attention to handle the probabilistic influence mode. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed MMSBR. Further analysis also proves that our MMSBR can alleviate the cold-start problem in SBR effectively.
Learning-based multi-view stereo (MVS) method heavily relies on feature matching, which requires distinctive and descriptive representations. An effective solution is to apply non-local feature aggregation, e.g., Transformer. Albeit useful, these techniques introduce heavy computation overheads for MVS. Each pixel densely attends to the whole image. In contrast, we propose to constrain non-local feature augmentation within a pair of lines: each point only attends the corresponding pair of epipolar lines. Our idea takes inspiration from the classic epipolar geometry, which shows that one point with different depth hypotheses will be projected to the epipolar line on the other view. This constraint reduces the 2D search space into the epipolar line in stereo matching. Similarly, this suggests that the matching of MVS is to distinguish a series of points lying on the same line. Inspired by this point-to-line search, we devise a line-to-point non-local augmentation strategy. We first devise an optimized searching algorithm to split the 2D feature maps into epipolar line pairs. Then, an Epipolar Transformer (ET) performs non-local feature augmentation among epipolar line pairs. We incorporate the ET into a learning-based MVS baseline, named ET-MVSNet. ET-MVSNet achieves state-of-the-art reconstruction performance on both the DTU and Tanks-and-Temples benchmark with high efficiency. Code is available at https://github.com/TQTQliu/ET-MVSNet.
We have devised a data-driven framework for uncovering hidden control strategies used by an evolutionary system described by an evolutionary probability distribution. This innovative framework enables deciphering of the concealed mechanisms that contribute to the progression or mitigation of such situations as the spread of COVID-19. Novel algorithms are used to estimate the optimal control in tandem with the parameters for evolution in general dynamical systems, thereby extending the concept of model predictive control. This is a significant departure from conventional control methods, which require knowledge of the system to manipulate its evolution and of the controller's strategy or parameters. We used a generalized additive model, supplemented by extensive statistical testing, to identify a set of predictor covariates closely linked to the control. Using real-world COVID-19 data, we successfully delineated the descriptive behaviors of the COVID-19 epidemics in five prefectures in Japan and nine countries. We compared these nine countries and grouped them on the basis of shared profiles, providing valuable insights into their pandemic responses. Our findings underscore the potential of our framework as a powerful tool for understanding and managing complex evolutionary processes.
This paper explores the internal structure of two quantum mechanics datasets (QM7b, QM9), composed of several thousands of organic molecules and described in terms of electronic properties. Understanding the structure and characteristics of this kind of data is important when predicting the atomic composition from the properties in inverse molecular designs. Intrinsic dimension analysis, clustering, and outlier detection methods were used in the study. They revealed that for both datasets the intrinsic dimensionality is several times smaller than the descriptive dimensions. The QM7b data is composed of well defined clusters related to atomic composition. The QM9 data consists of an outer region predominantly composed of outliers, and an inner core region that concentrates clustered, inliner objects. A significant relationship exists between the number of atoms in the molecule and its outlier/inner nature. Despite the structural differences, the predictability of variables of interest for inverse molecular design is high. This is exemplified with models estimating the number of atoms of the molecule from both the original properties, and from lower dimensional embedding spaces.
Social tipping points are promising levers to achieve net-zero greenhouse gas emission targets. They describe how social, political, economic or technological systems can move rapidly into a new state if cascading positive feedback mechanisms are triggered. Analysing the potential of social tipping for rapid decarbonization requires considering the inherent complexity of social systems. Here, we identify that existing scientific literature is inclined to a narrative-based account of social tipping, lacks a broad empirical framework and a multi-systems view. We subsequently outline a dynamic systems approach that entails (i) a systems outlook involving interconnected feedback mechanisms alongside cross-system and cross-scale interactions, and including a socioeconomic and environmental injustice perspective (ii) directed data collection efforts to provide empirical evidence for and monitor social tipping dynamics, (iii) global, integrated, descriptive modelling to project future dynamics and provide ex-ante evidence for interventions. Research on social tipping must be accordingly solidified for climate policy relevance.
"mdendro" is an R package that provides a comprehensive collection of linkage methods for agglomerative hierarchical clustering on a matrix of proximity data (distances or similarities), returning a multifurcated dendrogram or multidendrogram. Multidendrograms can group more than two clusters at the same time, solving the nonuniqueness problem that arises when there are ties in the data. This problem causes that different binary dendrograms are possible depending both on the order of the input data and on the criterion used to break ties. Weighted and unweighted versions of the most common linkage methods are included in the package, which also implements two parametric linkage methods. In addition, package "mdendro" provides five descriptive measures to analyze the resulting dendrograms: cophenetic correlation coefficient, space distortion ratio, agglomeration coefficient, chaining coefficient and tree balance.
Automated image caption generation is essential for improving the accessibility and understanding of visual content. In this study, we introduce FaceGemma, a model that accurately describes facial attributes such as emotions, expressions, and features. Using FaceAttdb data, we generated descriptions for 2000 faces with the Llama 3 - 70B model and fine-tuned the PaliGemma model with these descriptions. Based on the attributes and captions supplied in FaceAttDB, we created a new description dataset where each description perfectly depicts the human-annotated attributes, including key features like attractiveness, full lips, big nose, blond hair, brown hair, bushy eyebrows, eyeglasses, male, smile, and youth. This detailed approach ensures that the generated descriptions are closely aligned with the nuanced visual details present in the images. Our FaceGemma model leverages an innovative approach to image captioning by using annotated attributes, human-annotated captions, and prompt engineering to produce high-quality facial descriptions. Our method significantly improved caption quality, achieving an average BLEU-1 score of 0.364 and a METEOR score of 0.355. These metrics demonstrate the effectiveness of incorporating facial attributes into image captioning, providing more accurate and descriptive captions for portrait images.
Existing domain adaptation (DA) and generalization (DG) methods in object detection enforce feature alignment in the visual space but face challenges like object appearance variability and scene complexity, which make it difficult to distinguish between objects and achieve accurate detection. In this paper, we are the first to address the problem of semi-supervised domain generalization by exploring vision-language pre-training and enforcing feature alignment through the language space. We employ a novel Cross-Domain Descriptive Multi-Scale Learning (CDDMSL) aiming to maximize the agreement between descriptions of an image presented with different domain-specific characteristics in the embedding space. CDDMSL significantly outperforms existing methods, achieving 11.7% and 7.5% improvement in DG and DA settings, respectively. Comprehensive analysis and ablation studies confirm the effectiveness of our method, positioning CDDMSL as a promising approach for domain generalization in object detection tasks.
In the realm of research, the detection/recognition of text within images/videos captured by cameras constitutes a highly challenging problem for researchers. Despite certain advancements achieving high accuracy, current methods still require substantial improvements to be applicable in practical scenarios. Diverging from text detection in images/videos, this paper addresses the issue of text detection within license plates by amalgamating multiple frames of distinct perspectives. For each viewpoint, the proposed method extracts descriptive features characterizing the text components of the license plate, specifically corner points and area. Concretely, we present three viewpoints: view-1, view-2, and view-3, to identify the nearest neighboring components facilitating the restoration of text components from the same license plate line based on estimations of similarity levels and distance metrics. Subsequently, we employ the CnOCR method for text recognition within license plates. Experimental results on the self-collected dataset (PTITPlates), comprising pairs of images in various scenarios, and the publicly available Stanford Cars Dataset, demonstrate the superiority of the proposed method over existing approaches.
Nowadays, increasingly more data are available as knowledge graphs (KGs). While this data model supports advanced reasoning and querying, they remain difficult to mine due to their size and complexity. Graph mining approaches can be used to extract patterns from KGs. However this presents two main issues. First, graph mining approaches tend to extract too many patterns for a human analyst to interpret (pattern explosion). Second, real-life KGs tend to differ from the graphs usually treated in graph mining: they are multigraphs, their vertex degrees tend to follow a power-law, and the way in which they model knowledge can produce spurious patterns. Recently, a graph mining approach named GraphMDL+ has been proposed to tackle the problem of pattern explosion, using the Minimum Description Length (MDL) principle. However, GraphMDL+, like other graph mining approaches, is not suited for KGs without adaptations. In this paper we propose KG-MDL, a graph pattern mining approach based on the MDL principle that, given a KG, generates a human-sized and descriptive set of graph patterns, and so in a parameter-less and anytime way. We report on experiments on medium-sized KGs showing that our approach generates sets of patterns that are both small enough to be interpreted by humans and descriptive of the KG. We show that the extracted patterns highlight relevant characteristics of the data: both of the schema used to create the data, and of the concrete facts it contains. We also discuss the issues related to mining graph patterns on knowledge graphs, as opposed to other types of graph data.