Road change detection (RCD) is crucial for intelligent transportation, disaster assessment, and urban planning. However, current general change detection (CD) methods focus on various targets, such as buildings, while less attention is paid to the CD of narrow and elongated roads. Compared with general CD, RCD may still be limited by the following two aspects: On the one hand, the road usually occupies a small proportion of pixels in remote sensing images (RSIs) and is often easily blocked by buildings, trees, etc., making it difficult to ensure the integrity and connectivity of road structural features in RCD. On the other hand, RCD may easily be confused with the semantic information of similar material backgrounds (such as parking lots and building roofs) due to the lack of salient road features. To overcome the above limitations, we propose a skeleton-guided large-kernel attention Transformer (SLAFormer) for RCD, which can focus on salient road structural and semantic features to enhance its performance. In the proposed SLAFormer, we construct a novel skeleton-guided large-kernel attention module (SLKAM) and a frequency-guided cross spatial-channel difference module (FSCDM) to achieve the above goals. The SLKAM is used to make the model focus on road-specific skeleton features, which preserve the overall structure and morphology of roads to enhance the continuity and integrity of road features. In addition, the FSCDM is devised to better capture small-scale road changes and reduce semantic confusion with similar backgrounds, thereby enhancing change regions and extracting highly discriminative road difference information. Extensive experiments on two public RCD datasets show that ours achieves better RCD accuracy compared with several state-of-the-art (SOTA) approaches. The code will be available at https://github.com/TongfeiLiu/SLAFormer-for-RCD.

Vision-centric 3D environment understanding is both vi-tal and challenging for autonomous driving systems. Re-cently, object-free methods have attracted considerable at-tention. Such methods perceive the world by predicting the semantics of discrete voxel grids but fail to construct continuous and accurate obstacle surfaces. To this end, in this paper, we propose SurroundSDF to implicitly predict the signed distance field (SDF) and semantic field for the continuous perception from surround images. Specifically, we introduce a query-based approach and utilize SDF con-strained by the Eikonal formulation to accurately describe the surfaces of obstacles. Furthermore, considering the absence of precise SDF ground truth, we propose a novel weakly supervised paradigm for SDF, referred to as the Sandwich Eikonal formulation, which emphasizes applying correct and dense constraints on both sides of the surface, thereby enhancing the perceptual accuracy of the surface. Experiments suggest that our method achieves SOTA for both occupancy prediction and 3D scene reconstruction tasks on the nuScenes dataset.

Scene graph generation is a structured prediction task aiming to explicitly model objects and their relationships via constructing a visually-grounded scene graph for an input image. Currently, the message passing neural network based mean field variational Bayesian methodology is the ubiquitous solution for such a task, in which the variational inference objective is often assumed to be the classical evidence lower bound. However, the variational approximation inferred from such loose objective generally underestimates the underlying posterior, which often leads to inferior generation performance. In this paper, we propose a novel importance weighted structure learning method aiming to approximate the underlying log-partition function with a tighter importance weighted lower bound, which is computed from multiple samples drawn from a reparameterizable Gumbel-Softmax sampler. A generic entropic mirror descent algorithm is applied to solve the resulting constrained variational inference task. The proposed method achieves the state-of-the-art performance on various popular scene graph generation benchmarks.

Scene graph generation aims to interpret an input image by explicitly modelling the objects contained therein and their relationships. In existing methods the problem is predominantly solved by message passing neural network models. Unfortunately, in such models, the variational distributions generally ignore the structural dependencies among the output variables, and most of the scoring functions only consider pairwise dependencies. This can lead to inconsistent interpretations. In this article, we propose a novel neural belief propagation method seeking to replace the traditional mean field approximation with a structural Bethe approximation. To find a better bias-variance trade-off, higher-order dependencies among three or more output variables are also incorporated into the relevant scoring function. The proposed method achieves the state-of-the-art performance on various popular scene graph generation benchmarks.

As a structured prediction task, scene graph generation aims to build a visually-grounded scene graph to explicitly model objects and their relationships in an input image. Currently, the mean field variational Bayesian framework is the de facto methodology used by the existing methods, in which the unconstrained inference step is often implemented by a message passing neural network. However, such formulation fails to explore other inference strategies, and largely ignores the more general constrained optimization models. In this paper, we present a constrained structure learning method, for which an explicit constrained variational inference objective is proposed. Instead of applying the ubiquitous message-passing strategy, a generic constrained optimization method - entropic mirror descent - is utilized to solve the constrained variational inference step. We validate the proposed generic model on various popular scene graph generation benchmarks and show that it outperforms the state-of-the-art methods.

Event cameras open up new possibilities for robotic perception due to their low latency and high dynamic range. On the other hand, developing effective event-based vision algorithms that fully exploit the beneficial properties of event cameras remains work in progress. In this paper, we focus on event-based visual odometry (VO). While existing event-driven VO pipelines have adopted continuous-time representations to asynchronously process event data, they either assume a known map, restrict the camera to planar trajectories, or integrate other sensors into the system. Towards map-free event-only monocular VO in SE(3), we propose an asynchronous structure-from-motion optimisation back-end. Our formulation is underpinned by a principled joint optimisation problem involving non-parametric Gaussian Process motion modelling and incremental maximum a posteriori inference. A high-performance incremental computation engine is employed to reason about the camera trajectory with every incoming event. We demonstrate the robustness of our asynchronous back-end in comparison to frame-based methods which depend on accurate temporal accumulation of measurements.

A useful application of event sensing is visual odometry, especially in settings that require high-temporal resolution. The state-of-the-art method of contrast maximisation recovers the motion from a batch of events by maximising the contrast of the image of warped events. However, the cost scales with image resolution and the temporal resolution can be limited by the need for large batch sizes to yield sufficient structure in the contrast image1. In this work, we propose spatiotemporal registration as a compelling technique for event-based rotational motion estimation. We theoretically justify the approach and establish its fundamental and practical advantages over contrast maximisation. In particular, spatiotemporal registration also produces feature tracks as a by-product, which directly supports an efficient visual odometry pipeline with graph-based optimisation for motion averaging. The simplicity of our visual odometry pipeline allows it to process more than 1 M events/second. We also contribute a new event dataset for visual odometry, where motion sequences with large velocity variations were acquired using a high-precision robot arm2.

Visual semantic information comprises two important parts: the meaning of each visual semantic unit and the coherent visual semantic relation conveyed by these visual semantic units. Essentially, the former one is a visual perception task while the latter corresponds to visual context reasoning. Remarkable advances in visual perception have been achieved due to the success of deep learning. In contrast, visual semantic information pursuit, a visual scene semantic interpretation task combining visual perception and visual context reasoning, is still in its early stage. It is the core task of many different computer vision applications, such as object detection, visual semantic segmentation, visual relationship detection, or scene graph generation. Since it helps to enhance the accuracy and the consistency of the resulting interpretation, visual context reasoning is often incorporated with visual perception in current deep end-to-end visual semantic information pursuit methods. Surprisingly, a comprehensive review for this exciting area is still lacking. In this survey, we present a unified theoretical paradigm for all these methods, followed by an overview of the major developments and the future trends in each potential direction. The common benchmark datasets, the evaluation metrics and the comparisons of the corresponding methods are also introduced.

Contrast maximisation estimates the motion captured in an event stream by maximising the sharpness of the motion-compensated event image. To carry out contrast maximisation, many previous works employ iterative optimisation algorithms, such as conjugate gradient, which require good initialisation to avoid converging to bad local minima. To alleviate this weakness, we propose a new globally optimal event-based motion estimation algorithm. Based on branch-and-bound (BnB), our method solves rotational (3DoF) motion estimation on event streams, which supports practical applications such as video stabilisation and attitude estimation. Underpinning our method are novel bounding functions for contrast maximisation, whose theoretical validity is rigorously established. We show concrete examples from public datasets where globally optimal solutions are vital to the success of contrast maximisation. Despite its exact nature, our algorithm is currently able to process a 50,000-event input in ≈ 300 seconds (a locally optimal solver takes ≈ 30 seconds on the same input). The potentialfor GPU acceleration will also be discussed.

With the increasing popularity of social media and smart devices, the face as one of the key biometrics becomes vital for person identification. Among those face recognition algorithms, video-based face recognition methods could make use of both temporal and spatial information just as humans do to achieve better classification performance. However, they cannot identify individuals when certain key facial areas, such as eyes or nose, are disguised by heavy makeup or rubber/digital masks. To this end, we propose a novel deep spiking neural network architecture in this paper. It takes dynamic facial movements, the facial muscle changes induced by speaking or other activities, as the sole input. An event-driven continuous spike-timing-dependent plasticity learning rule with adaptive thresholding is applied to train the synaptic weights. The experiments on our proposed video-based disguise face database (MakeFace DB) demonstrate that the proposed learning method performs very well, i.e., it achieves from 95% to 100% correct classification rates under various realistic experimental scenarios.

Human beings can achieve reliable and fast visual pattern recognition with limited time and learning samples. Underlying this capability, ventral stream plays an important role in object representation and form recognition. Modeling the ventral steam may shed light on further understanding the visual brain in humans and building artificial vision systems for pattern recognition. The current methods to model the mechanism of ventral stream are far from exhibiting fast, continuous, and event-driven learning like the human brain. To create a visual system similar to ventral stream in human with fast learning capability, in this paper, we propose a new spiking neural system with an event-driven continuous spike timing dependent plasticity (STDP) learning method using specific spiking timing sequences. Two novel continuous input mechanisms have been used to obtain the continuous input spiking pattern sequence. With the event-driven STDP learning rule, the proposed learning procedure will be activated if the neuron receive one pre- or post-synaptic spike event. The experimental results on MNIST database show that the proposed method outperforms all other methods in fast learning scenarios and most of the current models in exhaustive learning experiments.

Face is a vital biometric for personal identification. However, the current video-based face recognition methods could not cope with large variances such as heavy makeups, disguised faces with rubber/digital masks or faces with certain areas (eyes, nose, or mouth) invisible. In this paper, we proposed a deep spiking neural network (SNN) architecture with the dynamic facial movements (facial muscle changes caused by speaking) as the sole input for the video-based disguise face recognition application. An event-driven continuous spike-timing dependent plasticity (STDP) learning algorithm with adaptive thresholding has been applied to train the synaptic weights. The proposed video-based disguise face recognition (VDFR) learning method achieves 95% correct classification rate on our proposed video-based disguise face database (MakeFace DB).

Neuroscience study shows mammalian brain only use millisecond scale time window to process complicated real-life recognition scenarios. However, such speed cannot be achieved by traditional rate-based spiking neural network (SNN). Compared with spiking rate, the specific spiking timing (also called spiking pattern) may convey much more information. In this paper, by using modified rank order coding scheme, the generated absolute analog features have been encoded into the first spike wave with specific spatiotemporal structural information. An intuitive yet powerful feed-forward spiking neural network framework has been proposed, along with its own unsupervised spike-timing-dependent plasticity (STDP) learning rule with dynamic post-synaptic potential threshold. Compared with other state-of-art spiking algorithms, the proposed method uses biologically plausible STDP learning method to learn the selectivity while the dynamic post-synaptic potential threshold guarantees no training sample will be ignored during the learning procedure. Furthermore, unlike the complicated frameworks used in those state-of-art spiking algorithms, the proposed intuitive spiking neural network is not time-consuming and quite capable of on-line learning. A satisfactory experimental result has been achieved on classic MNIST handwritten character database.