Objective. Spike sorting, a critical step in neural data processing, aims to classify spiking events from single electrode recordings based on different waveforms. This study aims to develop a novel online spike sorter, NeuSort, using neuromorphic models, with the ability to adaptively adjust to changes in neural signals, including waveform deformations and the appearance of new neurons. Approach. NeuSort leverages a neuromorphic model to emulate template-matching processes. This model incorporates plasticity learning mechanisms inspired by biological neural systems, facilitating real-time adjustments to online parameters. Results. Experimental findings demonstrate NeuSort's ability to track neuron activities amidst waveform deformations and identify new neurons in real-time. NeuSort excels in handling non-stationary neural signals, significantly enhancing its applicability for long-term spike sorting tasks. Moreover, its implementation on neuromorphic chips guarantees ultra-low energy consumption during computation. Significance. NeuSort caters to the demand for real-time spike sorting in brain-machine interfaces through a neuromorphic approach. Its unsupervised, automated spike sorting process makes it a plug-and-play solution for online spike sorting.
Acoustic sensors play an important role in autonomous underwater vehicles (AUVs). Sidescan sonar (SSS) detects a wide range and provides photo-realistic images in high resolution. However, SSS projects the 3D seafloor to 2D images, which are distorted by the AUV's altitude, target's range and sensor's resolution. As a result, the same physical area can show significant visual differences in SSS images from different survey lines, causing difficulties in tasks such as pixel correspondence and template matching. In this paper, a canonical transformation method consisting of intensity correction and slant range correction is proposed to decrease the above distortion. The intensity correction includes beam pattern correction and incident angle correction using three different Lambertian laws (cos, cos2, cot), whereas the slant range correction removes the nadir zone and projects the position of SSS elements into equally horizontally spaced, view-point independent bins. The proposed method is evaluated on real data collected by a HUGIN AUV, with manually-annotated pixel correspondence as ground truth reference. Experimental results on patch pairs compare similarity measures and keypoint descriptor matching. The results show that the canonical transformation can improve the patch similarity, as well as SIFT descriptor matching accuracy in different images where the same physical area was ensonified.
How can we segment varying numbers of objects where each specific object represents its own separate class? To make the problem even more realistic, how can we add and delete classes on the fly without retraining or fine-tuning? This is the case of robotic applications where no datasets of the objects exist or application that includes thousands of objects (E.g., in logistics) where it is impossible to train a single model to learn all of the objects. Most current research on object segmentation for robotic grasping focuses on class-level object segmentation (E.g., box, cup, bottle), closed sets (specific objects of a dataset; for example, YCB dataset), or deep learning-based template matching. In this work, we are interested in open sets where the number of classes is unknown, varying, and without pre-knowledge about the objects' types. We consider each specific object as its own separate class. Our goal is to develop an object detector that requires no fine-tuning and can add any object as a class just by capturing a few images of the object. Our main idea is to break the segmentation pipelines into two steps by combining unseen object segmentation networks cascaded by class-adaptive classifiers. We evaluate our class-adaptive object detector on unseen datasets and compare it to a trained Mask R-CNN on those datasets. The results show that the performance varies from practical to unsuitable depending on the environment setup and the objects being handled. The code is available in our DoUnseen library repository.
Because of the advantages of easy deployment, low cost and non-contact, computer vision-based structural displacement acquisition technique has received wide attention and research in recent years. However, the displacement field acquisition of large-scale structures is a challenging topic due to the contradiction of camera field of view and resolution. This paper presents a large-scale structural displacement field calculation framework with integrated computer vision and physical constraints using only one camera. Firstly, the full-field image of the large-scale structure is obtained by processing the multi-view image using image stitching technique; secondly, the full-field image is meshed and the node displacements are calculated using an improved template matching method; and finally, the non-node displacements are described using shape functions considering physical constraints. The developed framework was validated using a scaled bridge model and evaluated by the proposed evaluation index for displacement field calculation accuracy. This paper can provide an effective way to obtain displacement fields of large-scale structures efficiently and cost-effectively.
Rotational motion blur caused by the circular motion of the camera or/and object is common in life. Identifying objects from images affected by rotational motion blur is challenging because this image degradation severely impacts image quality. Therefore, it is meaningful to develop image invariant features under rotational motion blur and then use them in practical tasks, such as object classification and template matching. This paper proposes a novel method to generate image moment invariants under general rotational motion blur and provides some instances. Further, we achieve their invariance to similarity transform. To the best of our knowledge, this is the first time that moment invariants for rotational motion blur have been proposed in the literature. We conduct extensive experiments on various image datasets disturbed by similarity transform and rotational motion blur to test these invariants' numerical stability and robustness to image noise. We also demonstrate their performance in image classification and handwritten digit recognition. Current state-of-the-art blur moment invariants and deep neural networks are chosen for comparison. Our results show that the moment invariants proposed in this paper significantly outperform other features in various tasks.
Continuous monitoring of human vital signs using non-contact mmWave radars is attractive due to their ability to penetrate garments and operate under different lighting conditions. Unfortunately, most prior research requires subjects to stay at a fixed distance from radar sensors and to remain still during monitoring. These restrictions limit the applications of radar vital sign monitoring in real life scenarios. In this paper, we address these limitations and present "Pi-ViMo", a non-contact Physiology-inspired Robust Vital Sign Monitoring system, using mmWave radars. We first derive a multi-scattering point model for the human body, and introduce a coherent combining of multiple scatterings to enhance the quality of estimated chest-wall movements. It enables vital sign estimations of subjects at any location in a radar's field of view. We then propose a template matching method to extract human vital signs by adopting physical models of respiration and cardiac activities. The proposed method is capable to separate respiration and heartbeat in the presence of micro-level random body movements (RBM) when a subject is at any location within the field of view of a radar. Experiments in a radar testbed show average respiration rate errors of 6% and heart rate errors of 11.9% for the stationary subjects and average errors of 13.5% for respiration rate and 13.6% for heart rate for subjects under different RBMs.
Template matching is a fundamental task in computer vision and has been studied for decades. It plays an essential role in manufacturing industry for estimating the poses of different parts, facilitating downstream tasks such as robotic grasping. Existing methods fail when the template and source images have different modalities, cluttered backgrounds or weak textures. They also rarely consider geometric transformations via homographies, which commonly exist even for planar industrial parts. To tackle the challenges, we propose an accurate template matching method based on differentiable coarse-to-fine correspondence refinement. We use an edge-aware module to overcome the domain gap between the mask template and the grayscale image, allowing robust matching. An initial warp is estimated using coarse correspondences based on novel structure-aware information provided by transformers. This initial alignment is passed to a refinement network using references and aligned images to obtain sub-pixel level correspondences which are used to give the final geometric transformation. Extensive evaluation shows that our method is significantly better than state-of-the-art methods and baselines, providing good generalization ability and visually plausible results even on unseen real data.
The detection of hazardous terrain during the planetary landing of spacecraft plays a critical role in assuring vehicle safety and mission success. A cheap and effective way of detecting hazardous terrain is through the use of visual cameras, which ensure operational ability from atmospheric entry through touchdown. Plagued by resource constraints and limited computational power, traditional techniques for visual hazardous terrain detection focus on template matching and registration to pre-built hazard maps. Although successful on previous missions, this approach is restricted to the specificity of the templates and limited by the fidelity of the underlying hazard map, which both require extensive pre-flight cost and effort to obtain and develop. Terrestrial systems that perform a similar task in applications such as autonomous driving utilize state-of-the-art deep learning techniques to successfully localize and classify navigation hazards. Advancements in spacecraft co-processors aimed at accelerating deep learning inference enable the application of these methods in space for the first time. In this work, we introduce You Only Crash Once (YOCO), a deep learning-based visual hazardous terrain detection and classification technique for autonomous spacecraft planetary landings. Through the use of unsupervised domain adaptation we tailor YOCO for training by simulation, removing the need for real-world annotated data and expensive mission surveying phases. We further improve the transfer of representative terrain knowledge between simulation and the real world through visual similarity clustering. We demonstrate the utility of YOCO through a series of terrestrial and extraterrestrial simulation-to-real experiments and show substantial improvements toward the ability to both detect and accurately classify instances of planetary terrain.
Recent advances in machine learning and neuro-inspired systems enabled the increased interest in efficient pattern recognition at the edge. A wide variety of applications, such as near-sensor classification, require fast and low-power approaches for pattern matching through the use of associative memories and their more well-known implementation, Content Addressable Memories (CAMs). Towards addressing the need for low-power classification, this work showcases an RRAM-based analogue CAM (ACAM) intended for template matching applications, providing a low-power reconfigurable classification engine for the extreme edge. The circuit uses a low component count at 6T2R2M, comparable with the most compact existing cells of this type. In this work, we demonstrate a hardware prototype, built with commercial off-the-shelf (COTS) components for the MOSFET-based circuits, that implements rows of 6T2R2M employing TiOx-based RRAM devices developed in-house, showcasing competitive matching window configurability and definition. Furthermore, through simulations, we validate the performance of the proposed circuit by using a commercially available 180nm technology and in-house RRAM data-driven model to assess the energy dissipation, exhibiting 60 pJ per classification event.
We present an effective method for the matching of multimodal images. Accurate image matching is the basis of various applications, such as image registration and structure from motion. Conventional matching methods fail when handling noisy multimodal image pairs with severe scale change, rotation, and nonlinear intensity distortion (NID). Toward this need, we introduce an image pyramid strategy to tackle scale change. We put forward an accurate primary orientation estimation approach to reduce the effect of image rotation at any angle. We utilize multi-scale and multi-orientation image filtering results and a feature-to-template matching scheme to ensure effective and accurate matching under large NID. Integrating these improvements significantly increases noise, scale, rotation, and NID invariant capability. Our experimental results confirm the excellent ability to achieve high-quality matches across various multimodal images. The proposed method outperforms the mainstream multimodal image matching methods in qualitative and quantitative evaluations. Our implementation is available at https://github.com/Zhongli-Fan/NISR.
We propose an effective method for removing thermal vibrations that complicate the task of analyzing complex dynamics in atomistic simulation of condensed matter. Our method iteratively subtracts thermal noises or perturbations in atomic positions using a denoising score function trained on synthetically noised but otherwise perfect crystal lattices. The resulting denoised structures clearly reveal underlying crystal order while retaining disorder associated with crystal defects. Purely geometric, agnostic to interatomic potentials, and trained without inputs from explicit simulations, our denoiser can be applied to simulation data generated from vastly different interatomic interactions. The denoiser is shown to improve existing classification methods such as common neighbor analysis and polyhedral template matching, reaching perfect classification accuracy on a recent benchmark dataset of thermally perturbed structures up to the melting point. Demonstrated here in a wide variety of atomistic simulation contexts, the denoiser is general, robust, and readily extendable to delineate order from disorder in structurally and chemically complex materials.
Object location prior is critical for the standard 6D object pose estimation setting. The prior can be used to initialize the 3D object translation and facilitate 3D object rotation estimation. Unfortunately, the object detectors that are used for this purpose do not generalize to unseen objects. Therefore, existing 6D pose estimation methods for unseen objects either assume the ground-truth object location to be known or yield inaccurate results when it is unavailable. In this paper, we address this problem by developing a method, LocPoseNet, able to robustly learn location prior for unseen objects. Our method builds upon a template matching strategy, where we propose to distribute the reference kernels and convolve them with a query to efficiently compute multi-scale correlations. We then introduce a novel translation estimator, which decouples scale-aware and scale-robust features to predict different object location parameters. Our method outperforms existing works by a large margin on LINEMOD and GenMOP. We further construct a challenging synthetic dataset, which allows us to highlight the better robustness of our method to various noise sources. Our project website is at: https://sailor-z.github.io/projects/3DV2024_LocPoseNet.html.
Selection of appropriate template matching algorithms to run effectively on real-time low-cost systems is always major issue. This is due to unpredictable changes in image scene which often necessitate more sophisticated real-time algorithms to retain image consistency. Inefficiency of low cost auxiliary hardware and time limitations are the major constraints in using these sorts of algorithms. The real-time system introduced here copes with these problems utilising a fast running template matching algorithm, which makes use of best colour band selection. The system uses fast running real-time algorithms to achieve template matching and vehicle classification at about 4 frames /sec. on low-cost hardware. The colour image sequences have been taken by a fixed CCTV camera overlooking a busy multi-lane road
Convolution blocks serve as local feature extractors and are the key to success of the neural networks. To make local semantic feature embedding rather explicit, we reformulate convolution blocks as feature selection according to the best matching kernel. In this manner, we show that typical ResNet blocks indeed perform local feature embedding via template matching once batch normalization (BN) followed by a rectified linear unit (ReLU) is interpreted as arg-max optimizer. Following this perspective, we tailor a residual block that explicitly forces semantically meaningful local feature embedding through using label information. Specifically, we assign a feature vector to each local region according to the classes that the corresponding region matches. We evaluate our method on three popular benchmark datasets with several architectures for image classification and consistently show that our approach substantially improves the performance of the baseline architectures.
In this paper, new methods are considered to detect knee joint areas in bilateral PA fixed flexion knee X-ray images. The methods are of template matching type where the distance criterion is based on the negative normalized cross-correlation. The manual annotations are made on only one side of a single bilateral image when the templates are selected. The best matching patch search is formulated as an unconstrained continuous domain minimization problem. For the minimization problem different optimization methods are considered. The main method of the paper is a trainable optimizer where the method is taught to take zoomed and possibly rotated patches from its input images which look like the template. In the experiments, we compare the minimum values found by different optimization methods. We also look at some test images to examine the correspondence between the minimum value and how well the knee area is localized. It seems that making annotations only to a single image enables to detect knee joint areas quite precisely.
Active consumption of digital documents has yielded scope for research in various applications, including search. Traditionally, searching within a document has been cast as a text matching problem ignoring the rich layout and visual cues commonly present in structured documents, forms, etc. To that end, we ask a mostly unexplored question: "Can we search for other similar snippets present in a target document page given a single query instance of a document snippet?". We propose MONOMER to solve this as a one-shot snippet detection task. MONOMER fuses context from visual, textual, and spatial modalities of snippets and documents to find query snippet in target documents. We conduct extensive ablations and experiments showing MONOMER outperforms several baselines from one-shot object detection (BHRL), template matching, and document understanding (LayoutLMv3). Due to the scarcity of relevant data for the task at hand, we train MONOMER on programmatically generated data having many visually similar query snippets and target document pairs from two datasets - Flamingo Forms and PubLayNet. We also do a human study to validate the generated data.
We propose AccoMontage2, a system capable of doing full-length song harmonization and accompaniment arrangement based on a lead melody. Following AccoMontage, this study focuses on generating piano arrangements for popular/folk songs and it carries on the generalized template-based retrieval method. The novelties of this study are twofold. First, we invent a harmonization module (which AccoMontage does not have). This module generates structured and coherent full-length chord progression by optimizing and balancing three loss terms: a micro-level loss for note-wise dissonance, a meso-level loss for phrase-template matching, and a macro-level loss for full piece coherency. Second, we develop a graphical user interface which allows users to select different styles of chord progression and piano texture. Currently, chord progression styles include Pop, R&B, and Dark, while piano texture styles include several levels of voicing density and rhythmic complexity. Experimental results show that both our harmonization and arrangement results significantly outperform the baselines. Lastly, we release AccoMontage2 as an online application as well as the organized chord progression templates as a public dataset.
In this paper, we propose a novel template matching method with a white balancing adjustment, called N-white balancing, which was proposed for multi-illuminant scenes. To reduce the influence of lighting effects, N-white balancing is applied to images for multi-illumination color constancy, and then a template matching method is carried out by using adjusted images. In experiments, the effectiveness of the proposed method is demonstrated to be effective in object detection tasks under various illumination conditions.
The insertion of through-hole components is a difficult task. As the tolerances of the holes are very small, minor errors in the insertion will result in failures. These failures can damage components and will require manual intervention for recovery. Errors can occur both from imprecise object grasps and bent pins. Therefore, it is important that a system can accurately determine the object's position and reject components with bent pins. By utilizing the constraints inherent in the object grasp a method using template matching is able to obtain very precise pose estimates. Methods for pin-checking are also implemented, compared, and a successful method is shown. The set-up is performed automatically, with two novel contributions. A deep learning segmentation of the pins is performed and the inspection pose is found by simulation. From the inspection pose and the segmented pins, the templates for pose estimation and pin check are then generated. To train the deep learning method a dataset of segmented through-hole components is created. The network shows a 97.3 % accuracy on the test set. The pin-segmentation network is also tested on the insertion CAD models and successfully segment the pins. The complete system is tested on three different objects, and experiments show that the system is able to insert all objects successfully. Both by correcting in-hand grasp errors and rejecting objects with bent pins.
Diabetic Retinopathy (DR) is a complication of long-standing, unchecked diabetes and one of the leading causes of blindness in the world. This paper focuses on improved and robust methods to extract some of the features of DR, viz. Blood Vessels and Exudates. Blood vessels are segmented using multiple morphological and thresholding operations. For the segmentation of exudates, k-means clustering and contour detection on the original images are used. Extensive noise reduction is performed to remove false positives from the vessel segmentation algorithm's results. The localization of Optic Disc using k-means clustering and template matching is also performed. Lastly, this paper presents a Deep Convolutional Neural Network (DCNN) model with 14 Convolutional Layers and 2 Fully Connected Layers, for the automatic, binary diagnosis of DR. The vessel segmentation, optic disc localization and DCNN achieve accuracies of 95.93%, 98.77% and 75.73% respectively. The source code and pre-trained model are available https://github.com/Sohambasu07/DR_2021