Recent advances in large video-language models have displayed promising outcomes in video comprehension. Current approaches straightforwardly convert video into language tokens and employ large language models for multi-modal tasks. However, this method often leads to the generation of irrelevant content, commonly known as "hallucination", as the length of the text increases and the impact of the video diminishes. To address this problem, we propose Vista-LLaMA, a novel framework that maintains the consistent distance between all visual tokens and any language tokens, irrespective of the generated text length. Vista-LLaMA omits relative position encoding when determining attention weights between visual and text tokens, retaining the position encoding for text and text tokens. This amplifies the effect of visual tokens on text generation, especially when the relative distance is longer between visual and text tokens. The proposed attention mechanism significantly reduces the chance of producing irrelevant text related to the video content. Furthermore, we present a sequential visual projector that projects the current video frame into tokens of language space with the assistance of the previous frame. This approach not only captures the temporal relationship within the video, but also allows less visual tokens to encompass the entire video. Our approach significantly outperforms various previous methods (e.g., Video-ChatGPT, MovieChat) on four challenging open-ended video question answering benchmarks. We reach an accuracy of 60.7 on the zero-shot NExT-QA and 60.5 on the zero-shot MSRVTT-QA, setting a new state-of-the-art performance. This project is available at https://jinxxian.github.io/Vista-LLaMA.
Segmentation of Arabic manuscripts into lines of text and words is an important step to make recognition systems more efficient and accurate. The problem of segmentation into text lines is solved since there are carefully annotated dataset dedicated to this task. However, To the best of our knowledge, there are no dataset annotating the word position of Arabic texts. In this paper, we present a new dataset specifically designed for historical Arabic script in which we annotate position in word level.
As the exorbitant expense of labeling autopilot datasets and the growing trend of utilizing unlabeled data, semi-supervised segmentation on point clouds becomes increasingly imperative. Intuitively, finding out more ``unspoken words'' (i.e., latent instance information) beyond the label itself should be helpful to improve performance. In this paper, we discover two types of latent labels behind the displayed label embedded in LiDAR and image data. First, in the LiDAR Branch, we propose a novel augmentation, Cylinder-Mix, which is able to augment more yet reliable samples for training. Second, in the Image Branch, we propose the Instance Position-scale Learning (IPSL) Module to learn and fuse the information of instance position and scale, which is from a 2D pre-trained detector and a type of latent label obtained from 3D to 2D projection. Finally, the two latent labels are embedded into the multi-modal panoptic segmentation network. The ablation of the IPSL module demonstrates its robust adaptability, and the experiments evaluated on SemanticKITTI and nuScenes demonstrate that our model outperforms the state-of-the-art method, LaserMix.
Breast cancer ranks as the most prevalent form of cancer diagnosed in women, and diagnosis faces several challenges, a change in the size, shape and appearance of breasts, dense breast tissue, lumps or thickening in the breast especially if in only one breast, lumps and nodules in the breast. The major challenge that faces deep learning diagnosis of breast cancer was its shape, size and position non-uniformity especially malignant cancer. This work proposed a deep learning system that increased the accuracy of classification of breast cancer types from ultrasound images. It reaches 99.29% accuracy, exceeding other previous work. First, image processing was applied to in enhance the quality of input images. Second, the image segmentation was performed using U-Net architecture. Third, many features are extracted using Mobilenet. Finally, the accuracy of proposed system was evaluated.
We present a novel multimodal dataset developed by expert astronomers to automate the detection and localisation of multi-component extended radio galaxies and their corresponding infrared hosts. The dataset comprises 4,155 instances of galaxies in 2,800 images with both radio and infrared modalities. Each instance contains information on the extended radio galaxy class, its corresponding bounding box that encompasses all of its components, pixel-level segmentation mask, and the position of its corresponding infrared host galaxy. Our dataset is the first publicly accessible dataset that includes images from a highly sensitive radio telescope, infrared satellite, and instance-level annotations for their identification. We benchmark several object detection algorithms on the dataset and propose a novel multimodal approach to identify radio galaxies and the positions of infrared hosts simultaneously.
The concepts of stability and balance represent many critical problems faced by engineering today. The inverted pendulum on a cart is one such non-linear, unstable, multivariate system whose goal is to determine a suitable control action given to the cart such that it stabilizes the pendulum in an upright vertical position. This paper therefore, aims to design and study a highly robust MISO control structure using Linear Quadratic Regulation, Fuzzy logic and Neural Networks called Two-Stage LQR-based-ANFIS (referred to as TS-LA) for the stabilization of Inverted Pendulums. The proposed controller is implemented on a Simulink model of the Inverted Pendulum constructed through relevant mathematical and state space modelling using Newtonian and Lagrangian mechanics. Applying external disturbances, transient parameters are obtained and are benchmarked against standard conventional controllers to perform comparative analysis and showcase its disturbance rejection capabilities.
In a developing embryo, information about the position of cells is encoded in the concentrations of "morphogen" molecules. In the fruit fly, the local concentrations of just a handful of proteins encoded by the gap genes are sufficient to specify position with a precision comparable to the spacing between cells along the anterior--posterior axis. This matches the precision of downstream events such as the striped patterns of expression in the pair-rule genes, but is not quite sufficient to define unique identities for individual cells. We demonstrate theoretically that this information gap can be bridged if positional errors are spatially correlated, with relatively long correlation lengths. We then show experimentally that these correlations are present, with the required strength, in the fluctuating positions of the pair-rule stripes, and this can be traced back to the gap genes. Taking account of these correlations, the available information matches the information needed for unique cellular specification, within error bars of ~2%. These observation support a precisionist view of information flow through the underlying genetic networks, in which accurate signals are available from the start and preserved as they are transformed into the final spatial patterns.
A control technique is developed via Reinforcement Learning that allows arbitrary controlling of the position of an acoustic cavitation bubble in a dual-frequency standing acoustic wave field. The agent must choose the optimal pressure amplitude values to manipulate the bubble position in the range of $x/\lambda_0\in[0.05, 0.25]$. To train the agent an actor-critic off-policy algorithm (Deep Deterministic Policy Gradient) was used that supports continuous action space, which allows setting the pressure amplitude values continuously within $0$ and $1\, \mathrm{bar}$. A shaped reward function is formulated that minimizes the distance between the bubble and the target position and implicitly encourages the agent to perform the position control within the shortest amount of time. In some cases, the optimal control can be 7 times faster than the solution expected from the linear theory.
Graph Transformers (GTs) facilitate the comprehension of graph-structured data by calculating the self-attention of node pairs without considering node position information. To address this limitation, we introduce an innovative and efficient framework that introduces Positional Encodings (PEs) into the Transformer, generating a set of learnable positional encodings in the hyperbolic space, a non-Euclidean domain. This approach empowers us to explore diverse options for optimal selection of PEs for specific downstream tasks, leveraging hyperbolic neural networks or hyperbolic graph convolutional networks. Additionally, we repurpose these positional encodings to mitigate the impact of over-smoothing in deep Graph Neural Networks (GNNs). Comprehensive experiments on molecular benchmark datasets, co-author, and co-purchase networks substantiate the effectiveness of hyperbolic positional encodings in enhancing the performance of deep GNNs.
We propose a novel Attentional Scale Sequence Fusion based You Only Look Once (YOLO) framework (ASF-YOLO) which combines spatial and scale features for accurate and fast cell instance segmentation. Built on the YOLO segmentation framework, we employ the Scale Sequence Feature Fusion (SSFF) module to enhance the multi-scale information extraction capability of the network, and the Triple Feature Encoder (TFE) module to fuse feature maps of different scales to increase detailed information. We further introduce a Channel and Position Attention Mechanism (CPAM) to integrate both the SSFF and TPE modules, which focus on informative channels and spatial position-related small objects for improved detection and segmentation performance. Experimental validations on two cell datasets show remarkable segmentation accuracy and speed of the proposed ASF-YOLO model. It achieves a box mAP of 0.91, mask mAP of 0.887, and an inference speed of 47.3 FPS on the 2018 Data Science Bowl dataset, outperforming the state-of-the-art methods. The source code is available at https://github.com/mkang315/ASF-YOLO.
3D instance segmentation plays a crucial role in comprehending 3D scenes. Despite recent advancements in this field, existing approaches exhibit certain limitations. These methods often rely on fixed instance positions obtained from sampled representative points in vast 3D point clouds, using center prediction or farthest point sampling. However, these selected positions may deviate from actual instance centers, posing challenges in precisely grouping instances. Moreover, the common practice of grouping candidate instances from a single type of coordinates introduces difficulties in identifying neighboring instances or incorporating edge points. To tackle these issues, we present a novel Transformer-based architecture, EipFormer, which comprises progressive aggregation and dual position embedding. The progressive aggregation mechanism leverages instance positions to refine instance proposals. It enhances the initial instance positions through weighted farthest point sampling and further refines the instance positions and proposals using aggregation averaging and center matching. Additionally, dual position embedding superposes the original and centralized position embeddings, thereby enhancing the model performance in distinguishing adjacent instances. Extensive experiments on popular datasets demonstrate that EipFormer achieves superior or comparable performance compared to state-of-the-art approaches.
Detecting spoofing attacks on the positions of unmanned aerial vehicles (UAVs) within a swarm is challenging. Traditional methods relying solely on individually reported positions and pairwise distance measurements are ineffective in identifying the misbehavior of malicious UAVs. This paper presents a novel systematic structure designed to detect and mitigate spoofing attacks in UAV swarms. We formulate the problem of detecting malicious UAVs as a localization feasibility problem, leveraging the reported positions and distance measurements. To address this problem, we develop a semidefinite relaxation (SDR) approach, which reformulates the non-convex localization problem into a convex and tractable semidefinite program (SDP). Additionally, we propose two innovative algorithms that leverage the proximity of neighboring UAVs to identify malicious UAVs effectively. Simulations demonstrate the superior performance of our proposed approaches compared to existing benchmarks. Our methods exhibit robustness across various swarm networks, showcasing their effectiveness in detecting and mitigating spoofing attacks. {\blue Specifically, the detection success rate is improved by up to 65\%, 55\%, and 51\% against distributed, collusion, and mixed attacks, respectively, compared to the benchmarks.
In this paper, we democratise 3D content creation, enabling precise generation of 3D shapes from abstract sketches while overcoming limitations tied to drawing skills. We introduce a novel part-level modelling and alignment framework that facilitates abstraction modelling and cross-modal correspondence. Leveraging the same part-level decoder, our approach seamlessly extends to sketch modelling by establishing correspondence between CLIPasso edgemaps and projected 3D part regions, eliminating the need for a dataset pairing human sketches and 3D shapes. Additionally, our method introduces a seamless in-position editing process as a byproduct of cross-modal part-aligned modelling. Operating in a low-dimensional implicit space, our approach significantly reduces computational demands and processing time.
Visual interactivity understanding within visual scenes presents a significant challenge in computer vision. Existing methods focus on complex interactivities while leveraging a simple relationship model. These methods, however, struggle with a diversity of appearance, situation, position, interaction, and relation in videos. This limitation hinders the ability to fully comprehend the interplay within the complex visual dynamics of subjects. In this paper, we delve into interactivities understanding within visual content by deriving scene graph representations from dense interactivities among humans and objects. To achieve this goal, we first present a new dataset containing Appearance-Situation-Position-Interaction-Relation predicates, named ASPIRe, offering an extensive collection of videos marked by a wide range of interactivities. Then, we propose a new approach named Hierarchical Interlacement Graph (HIG), which leverages a unified layer and graph within a hierarchical structure to provide deep insights into scene changes across five distinct tasks. Our approach demonstrates superior performance to other methods through extensive experiments conducted in various scenarios.
Deep neural networks (DNNs) are vulnerable to various types of adversarial examples, bringing huge threats to security-critical applications. Among these, adversarial patches have drawn increasing attention due to their good applicability to fool DNNs in the physical world. However, existing works often generate patches with meaningless noise or patterns, making it conspicuous to humans. To address this issue, we explore how to generate visually realistic adversarial patches to fool DNNs. Firstly, we analyze that a high-quality adversarial patch should be realistic, position irrelevant, and printable to be deployed in the physical world. Based on this analysis, we propose an effective attack called VRAP, to generate visually realistic adversarial patches. Specifically, VRAP constrains the patch in the neighborhood of a real image to ensure the visual reality, optimizes the patch at the poorest position for position irrelevance, and adopts Total Variance loss as well as gamma transformation to make the generated patch printable without losing information. Empirical evaluations on the ImageNet dataset demonstrate that the proposed VRAP exhibits outstanding attack performance in the digital world. Moreover, the generated adversarial patches can be disguised as the scrawl or logo in the physical world to fool the deep models without being detected, bringing significant threats to DNNs-enabled applications.
Salient object detection (SOD) in optical remote sensing images (ORSIs) has become increasingly popular recently. Due to the characteristics of ORSIs, ORSI-SOD is full of challenges, such as multiple objects, small objects, low illuminations, and irregular shapes. To address these challenges, we propose a concise yet effective Texture-Semantic Collaboration Network (TSCNet) to explore the collaboration of texture cues and semantic cues for ORSI-SOD. Specifically, TSCNet is based on the generic encoder-decoder structure. In addition to the encoder and decoder, TSCNet includes a vital Texture-Semantic Collaboration Module (TSCM), which performs valuable feature modulation and interaction on basic features extracted from the encoder. The main idea of our TSCM is to make full use of the texture features at the lowest level and the semantic features at the highest level to achieve the expression enhancement of salient regions on features. In the TSCM, we first enhance the position of potential salient regions using semantic features. Then, we render and restore the object details using the texture features. Meanwhile, we also perceive regions of various scales, and construct interactions between different regions. Thanks to the perfect combination of TSCM and generic structure, our TSCNet can take care of both the position and details of salient objects, effectively handling various scenes. Extensive experiments on three datasets demonstrate that our TSCNet achieves competitive performance compared to 14 state-of-the-art methods. The code and results of our method are available at https://github.com/MathLee/TSCNet.
There has been huge recent interest in the potential of making operational weather forecasts using machine learning techniques. As they become a part of the weather forecasting toolbox, there is a pressing need to understand how well current machine learning models can simulate high-impact weather events. We compare forecasts of Storm Ciar\'an, a European windstorm that caused sixteen deaths and extensive damage in Northern Europe, made by machine learning and numerical weather prediction models. The four machine learning models considered (FourCastNet, Pangu-Weather, GraphCast and FourCastNet-v2) produce forecasts that accurately capture the synoptic-scale structure of the cyclone including the position of the cloud head, shape of the warm sector and location of warm conveyor belt jet, and the large-scale dynamical drivers important for the rapid storm development such as the position of the storm relative to the upper-level jet exit. However, their ability to resolve the more detailed structures important for issuing weather warnings is more mixed. All of the machine learning models underestimate the peak amplitude of winds associated with the storm, only some machine learning models resolve the warm core seclusion and none of the machine learning models capture the sharp bent-back warm frontal gradient. Our study shows there is a great deal about the performance and properties of machine learning weather forecasts that can be derived from case studies of high-impact weather events such as Storm Ciar\'an.
Automatically determining the position of every slice in a CT scan is a basic yet powerful capability allowing fast retrieval of region of interest for visual inspection and automated analysis. Unlike conventional localization approaches which work at the slice level, we directly localize only a fraction of the slices and and then fit a linear model which maps slice index to its estimated axial anatomical position based on those slices. The model is then used to assign axial position to every slices of the scan. This approach proves to be both computationally efficient, with a typical processing time of less than a second per scan (regardless of its size), accurate, with a typical median localization error of 1 cm, and robust to different noise sources, imaging protocols, metal induced artifacts, anatomical deformations etc. Another key element of our approach is the introduction of a mapping confidence score. This score acts as a fail safe mechanism which allows a rejection of unreliable localization results in rare cases of anomalous scans. Our algorithm sets new State Of The Art results in terms of localization accuracy. It also offers a decrease of two orders of magnitude in processing time with respect to all published processing times. It was designed to be invariant to various scan resolutions, scan protocols, patient orientations, strong artifacts and various deformations and abnormalities. Additionally, our algorithm is the first one to the best of our knowledge which supports the entire body from head to feet and is not confined to specific anatomical region. This algorithm was tested on thousands of scans and proves to be very reliable and useful as a preprocessing stage for many applications.
It is widely believed that sparse supervision is worse than dense supervision in the field of depth completion, but the underlying reasons for this are rarely discussed. To this end, we revisit the task of radar-camera depth completion and present a new method with sparse LiDAR supervision to outperform previous dense LiDAR supervision methods in both accuracy and speed. Specifically, when trained by sparse LiDAR supervision, depth completion models usually output depth maps containing significant stripe-like artifacts. We find that such a phenomenon is caused by the implicitly learned positional distribution pattern from sparse LiDAR supervision, termed as LiDAR Distribution Leakage (LDL) in this paper. Based on such understanding, we present a novel Disruption-Compensation radar-camera depth completion framework to address this issue. The Disruption part aims to deliberately disrupt the learning of LiDAR distribution from sparse supervision, while the Compensation part aims to leverage 3D spatial and 2D semantic information to compensate for the information loss of previous disruptions. Extensive experimental results demonstrate that by reducing the impact of LDL, our framework with sparse supervision outperforms the state-of-the-art dense supervision methods with 11.6% improvement in Mean Absolute Error (MAE)} and 1.6x speedup in Frame Per Second (FPS)}. The code is available at https://github.com/megvii-research/Sparse-Beats-Dense.
As it is empirically observed that Vision Transformers (ViTs) are quite insensitive to the order of input tokens, the need for an appropriate self-supervised pretext task that enhances the location awareness of ViTs is becoming evident. To address this, we present DropPos, a novel pretext task designed to reconstruct Dropped Positions. The formulation of DropPos is simple: we first drop a large random subset of positional embeddings and then the model classifies the actual position for each non-overlapping patch among all possible positions solely based on their visual appearance. To avoid trivial solutions, we increase the difficulty of this task by keeping only a subset of patches visible. Additionally, considering there may be different patches with similar visual appearances, we propose position smoothing and attentive reconstruction strategies to relax this classification problem, since it is not necessary to reconstruct their exact positions in these cases. Empirical evaluations of DropPos show strong capabilities. DropPos outperforms supervised pre-training and achieves competitive results compared with state-of-the-art self-supervised alternatives on a wide range of downstream benchmarks. This suggests that explicitly encouraging spatial reasoning abilities, as DropPos does, indeed contributes to the improved location awareness of ViTs. The code is publicly available at https://github.com/Haochen-Wang409/DropPos.