Predicting human mobility remains a fundamental challenge, especially when individuals deviate from routine patterns due to exploration, disruptions, or rare events. While sequential models like Transformers excel at capturing regular movement patterns, their performance often degrades under nonroutine scenarios involving rare or unfamiliar transitions. To address this, we propose ROAM (Routine-Oriented Adaptive Mobility Predictor), a novel framework that jointly models human mobility from both sequential and relational perspectives, enabling adaptive handling of both routine and nonroutine behaviors during prediction. ROAM combines a sequential encoder that captures historically frequent transitions with a complementary graph-based relational reasoning module that encodes both user-specific and group-level mobility structures. To dynamically integrate these views, we introduce a hierarchical confidence-aware gating mechanism that adaptively balances sequential and relational predictions based on their internal reliability. Extensive experiments on real-world mobility datasets show that ROAM consistently outperforms state-of-the-art baselines in next location prediction. Further analysis reveals that the combination of sequential and relational reasoning substantially improves robustness, particularly under out-of-routine scenarios.

Accurate dam displacement prediction is vital for optimizing maintenance and ensuring structural safety. Nevertheless, current models often struggle to effectively capture the complex relationships between structural responses and environmental variables, alongside the interactions between temporal dynamics and multivariate data, resulting in suboptimal predictive accuracy. Therefore, we propose a dual-branch interactive fusion network (DBIFN) for dam displacement prediction using parallel temporal representation and gated cross-attention. The dual-branch architecture, which parallelly integrates the enhanced Transformer (eTransformer) and long short-term memory (LSTM), is designed to optimize feature extraction and interaction modeling across multiple dimensions. Specifically, eTransformer is dedicated to extracting features from targeted displacement sequences, while LSTM effectively processes auxiliary environmental dynamics, enabling a comprehensive analysis of underlying patterns within monitoring data. To fully fuse the interpreted temporal features from dual-branch outputs, we introduce a new cross-attention module to utilize the multi-dimensional gated attention unit to efficiently encode them into semantic representations, followed by a Kolmogorov-Arnold network mapping for further representation enhancement. The effectiveness of the proposed model is validated using real-world monitoring datasets collected from a concrete dam project, with experiments conducted across multiple monitoring points. Results demonstrate that DBIFN achieves superior prediction accuracy compared to both single-branch and conventional baseline models. Across all monitoring points, the proposed model can effectively capture temporal variations, attaining an average coefficient of determination of over 0.95 on the test set and outperforming comparative models in most metrics. Furthermore, statistical significance testing confirms the reliability and reproducibility of the results, while computational efficiency is maintained within inference time constraints. These findings offer valuable insights into the practical application of DBIFN-based monitoring models and support informed decision-making.

Training embodied agents to perform complex robotic tasks presents significant challenges due to the entangled factors of task compositionality, environmental diversity, and dynamic changes. In this work, we introduce a novel imitation learning framework to train closed-loop concept-guided policies that enhance long-horizon task performance by leveraging discovered manipulation concepts. Unlike methods that rely on predefined skills and human-annotated labels, our approach allows agents to autonomously abstract manipulation concepts from their proprioceptive states, thereby alleviating misalignment due to ambiguities in human semantics and environmental complexity. Our framework comprises two primary components: an *Automatic Concept Discovery* module that identifies meaningful and consistent manipulation concepts, and a *Concept-Guided Policy Learning* module that effectively utilizes these manipulation concepts for adaptive task execution, including a *Concept Selection Transformer* for concept-based guidance and a *Concept-Guided Policy* for action prediction with the selected concepts. Experiments demonstrate that our approach significantly outperforms baseline methods across a range of tasks and environments, while showcasing emergent consistency in motion patterns associated with the discovered manipulation concepts. Codes are available at: https://github.com/PeiZhou26/AutoCGP.

The primary technical challenge for multi-input multi-output synthetic aperture radar (MIMO-SAR) systems is separating independent channel responses from aliased echoes while maintaining imaging performance. However, the most promising short-term shift-orthogonal (STSO) and segmented-phase-code (SPC) waveform require the use of elevation digital beamforming (DBF) to achieve echo separation. The cost of using elevation DBF for echo separation is the loss of elevation degrees of freedom and a significant increase in system complexity. To solve this problem, this paper proposes a novel coded waveform that introduces phase characteristics for echo separation through two-dimensional phase encoding of the transmitted waveform in both elevation and inter-pulse (azimuth) direction. In this scheme, azimuth DBF is used in the Doppler frequency domain to suppress interference signals, while elevation phase demodulation is employed to separate the echoes. This scheme eliminates the dependence of MIMO-SAR on waveform orthogonality and allows the direct use of a large number of single-station waveforms, providing flexibility in waveform selection. Additionally, retaining more degrees of freedom enables the multi-modal operation of MIMO-SAR. Finally, detailed simulation experiments are performed to verify the potential of the proposed scheme, and advantages and contributions are systematically analyzed.

We focus on the self-supervised discovery of manipulation concepts that can be adapted and reassembled to address various robotic tasks. We propose that the decision to conceptualize a physical procedure should not depend on how we name it (semantics) but rather on the significance of the informativeness in its representation regarding the low-level physical state and state changes. We model manipulation concepts -- discrete symbols -- as generative and discriminative goals and derive metrics that can autonomously link them to meaningful sub-trajectories from noisy, unlabeled demonstrations. Specifically, we employ a trainable codebook containing encodings --symbols -- capable of synthesizing the end-state of a sub-trajectory given the current state (generative informativeness). Moreover, the encoding corresponding to a particular sub-trajectory should differentiate the state within and outside it and confidently predict the subsequent action based on the gradient of its discriminative score (discriminative informativeness). These metrics, which do not rely on human annotation, can be seamlessly integrated into a VQ-VAE framework, enabling the partitioning of demonstrations into semantically consistent sub-trajectories, fulfilling the purpose of discovering manipulation concepts and the corresponding (sub)-goal states. We evaluate the effectiveness of the learned concepts by training policies that utilize them as guidance, demonstrating superior performance compared to other baselines. Additionally, our discovered manipulation concepts compare favorably to human-annotated ones, while saving much manual effort. The code and trained models will be made public.

Typically, synthetic aperture radar (SAR) utilizes linear frequency modulation (LFM) signal to acquire high-resolution images, requiring spectral windowing to suppress sidelobes while sacrificing signal-to-noise ratio (SNR). In contrast to LFM signal, nonlinear frequency modulation (NLFM) signal can reconstruct the signal power spectral density (PSD) without sacrificing SNR, providing autocorrelation outputs with lower sidelobes. Despite the excellent application potential of NLFM signal, the real-time generation of NLFM faces numerous challenges due to the high complexity of the systems involved and constraints imposed by waveform generator devices. In this letter, a low-complexity, high-precision and high-resolution intermediate-frequency NLFM signal generation device is developed, requiring only eleven parameters to generate real-time NLFM signal of arbitrary time width and bandwidth, with a maximum bandwidth reaching 1.2 GHz. This NLFM signal generator will be employed in the unmanned aerial vehicle (UAV) SAR system. Finally, the performance of the NLFM signal generator has been validated through ground experimental results.

Recent years have witnessed the rapid development of deep learning in various aspects, such as image classification, face recognition, and object detection. Yet, these approaches often focus on a single entity. The relationship between different entities is remained to be explored. Cohabitation is a kind of important relationship. In a scenario of residential entries, knowing the relationship of cohabitation could a) prevent tailgaters; b) identify unregistered strangers, and especially c) prevent the disease from spreading during the Cov-19 period. In this paper, we propose a method combining computer vision with graph algorithms to discover the cohabitation relationships as long-term and regular co-occurrence. We demonstrate the method beneficial to both industrial and technical aspects.

Existing enhancement methods are empirically expected to help the high-level end computer vision task: however, that is observed to not always be the case in practice. We focus on object or face detection in poor visibility enhancements caused by bad weathers (haze, rain) and low light conditions. To provide a more thorough examination and fair comparison, we introduce three benchmark sets collected in real-world hazy, rainy, and low-light conditions, respectively, with annotated objects/faces. We launched the UG2+ challenge Track 2 competition in IEEE CVPR 2019, aiming to evoke a comprehensive discussion and exploration about whether and how low-level vision techniques can benefit the high-level automatic visual recognition in various scenarios. To our best knowledge, this is the first and currently largest effort of its kind. Baseline results by cascading existing enhancement and detection models are reported, indicating the highly challenging nature of our new data as well as the large room for further technical innovations. Thanks to a large participation from the research community, we are able to analyze representative team solutions, striving to better identify the strengths and limitations of existing mindsets as well as the future directions.

Web has become increasingly oriented to multimedia content. Most information on the web is conveyed from images. Text localization in web image plays an important role in web image information extraction and retrieval. Current works on text localization in web images assume that text regions are in homogenous color and high contrast. Hence, the approaches may fail when text regions are in multi-color or imposed in complex background. In this paper, we propose a text extraction algorithm from web images based on the probabilistic candidate selection model. The model firstly segments text region candidates from input images using wavelet, Gaussian mixture model (GMM) and triangulation. The likelihood of a candidate region containing text is then learnt using a Bayesian probabilistic model from two features, namely, histogram of oriented gradient (HOG) and local binary pattern histogram Fourier feature (LBP-HF). Finally best candidate regions are integrated to form text regions. The algorithm is evaluated using 155 non-homogenous web images containing around 600 text regions. The results show that the proposed model is able to extract text regions from non-homogenous images effectively.

Computed tomography (CT) is used widely in traumatic brain injury diagnosis. One axial brain CT scan consists of multiple slices with different heights along the brain axial direction. Indexing of brain CT slices is to order the slices and align each individual slice onto the corresponding brain axial height, which is an important step in content-based image retrieval and computer-assisted diagnosis. Current existing methods for this indexing task are through the image registration techniques by registering 2D image slices onto a 3D brain atlas. In this paper, instead of using the registration methods, we propose a fast indexing method using anatomical feature classification. In our method, the brain CT scan is divided into 6 height levels along the axial direction so that slices in each level share similar anatomical structure. In this way, the indexing problem becomes a classification problem that one series of scan slices are to be classified into 6 classes. Experimental results show that the proposed method is effective and efficient.

In intracranial pathological examinations using CT scan, brain midline shift (MLS) is an important diagnostic feature indicating the pathological severity and patient's survival possibility. In this paper, we develop a new method of tracing the brain midline shift in traumatic brain injury (TBI) CT images using its original cause - the hemorrhage. Firstly, we model the relationship between the hemorrhage and the midline deformation caused by it using a linear regression model (H-MLS model). Secondly, using the H-MLS model, the deformed midline is predicted from the hemorrhage detected in CT images. Finally, the predicted deformed midline is adjusted according to the visual symmetry information. Preliminary experiments show that the proposed method is effective and time-efficient.

Multi-slice computer tomography (CT) scans are widely used in todaypsilas diagnosis of head traumas. It is effective to disclose the bleeding and fractures. In this paper, we present an automated detection of CT scan slices which contain hemorrhages. Our method is robust towards various rotation, displacement and motion blur. Detection of these pathological slices will be useful for further diagnosis and retrieval.

Graphical components information extraction is a crucial step in the chart recognition and understanding process. However, existing methods of information extraction from chart images either are type-dependent or rely on certain assumptions. In this paper, we present a general method to extract vectorized graphical information from scientific chart images. Our algorithm firstly constructs a data structure called directional single-connected chains (DSCC). It then employs ellipse-specific fitting and orthogonal diagonalization to calculate the curvatures of the chains and classify the chains into either straight lines or arcs. Finally we combine all straight lines and all arcs accordingly and use linear regression to compute their attributes. The DSCC has a good property in that it is less susceptible to noise. The experiment results show that our algorithm is efficient, robust and accurate.