Abstract Crowd evacuation has gained increasing attention in recent years. The agent‐based method has shown a superior capability to simulate complex behaviors during crowd evacuation simulation. For agent modeling, most existing methods only consider the decision process but ignore the detailed physical motion. In this article, we propose a hierarchical framework for crowd evacuation simulation, which combines the agent decision model with the agent motion model. In the decision model, we integrate emotional contagion and scene information to determine global path planning and local collision avoidance. In the motion model, we introduce a physics‐based character control method and control agent motion using deep reinforcement learning. Based on the decision strategy, the decision model can use a signal to control the agent motion in the motion model. Compared with existing methods, our framework can simulate physical interactions between agents and the environment. The results of the crowd evacuation simulation demonstrate that our framework can simulate crowd evacuation with physical fidelity.

Spike camera is a novel bio-inspired vision sensor that mimics the sampling mechanism of the primate fovea. It presents high temporal resolution and dynamic range, showing great potentials in the high-speed moving object recognition task, which has not been fully explored in the Multimedia community due to the lack of data and annotations. This paper contributes the first large-scale High-Speed Spiking Recognition (HSSR) dataset, by recording high-speed moving objects using a spike camera. The HSSR dataset contains 135,000 indoor objects annotated using ImageNet labels and 3,100 outdoor objects collected from real-world scenarios. Furthermore, we propose an original spiking recognition framework, which employs long-term spike stream features to supervise the feature learning from short-term spike streams. This framework improves the recognition accuracy, meanwhile substantially decreasing the recognition latency, making our method can accurately recognize moving objects at an equivalent speed of 514 km/h, using only 1 ms of spike stream. Experimental results show that, the proposed method achieves 76.5% accuracy for recognizing 100 fine-grained indoor objects and 84.3% accuracy for recognizing 8 outdoor objects using 1 ms of spike streams. Resources will be available at https://github.com/Evin-X/HSSR.

It is challenging to bridge the performance gap between binary convolutional neural network (BCNN) and floating-point CNN (FCNN). This performance gap is mainly caused by the inferior modeling capability and training strategy of BCNN, which leads to substantial residuals in intermediate feature maps between BCNN and FCNN. To minimize the performance gap, we enforce BCNN to produce similar intermediate feature maps with the ones of FCNN. This intuition leads to a more effective training strategy for BCNN, i.e., optimizing each binary convolutional block with blockwise distillation loss derived from FCNN. The goal of minimizing the residuals in intermediate feature maps also motivates us to update the binary convolutional block architecture to facilitate the optimization of blockwise distillation loss. Specifically, a lightweight shortcut branch is inserted into each binary convolutional block to complement residuals at each block. Benefited from its squeeze-and-interaction (SI) structure, this shortcut branch introduces a fraction of parameters, e.g., less than 10% overheads, but effectively boosts the modeling capability of binary convolution blocks in BCNN. Extensive experiments on ImageNet demonstrate the superior performance of our method in both classification efficiency and accuracy, e.g., BCNN trained with our methods achieves the accuracy of 60.45% on ImageNet, better than many state-of-the-art ones.

Neural network pruning and quantization are two major lines of network compression. This raises a natural question that whether we can find the optimal compression by considering multiple network compression criteria in a unified framework. This paper incorporates two criteria and seeks layer-wise compression by leveraging the meta-learning framework. A regularization loss is applied to unify the constraint of input and output channel numbers, bit-width of network activations and weights, so that the compressed network can satisfy a given Bit-OPerations counts (BOPs) constraint. We further propose an iterative compression constraint for optimizing the compression procedure, which effectively achieves a high compression rate and maintains the original network performance. Extensive experiments on various networks and vision tasks show that the proposed method yields better performance and compression rates than recent methods. For instance, our method achieves better image classification accuracy and compactness than the recent DJPQ. It achieves similar performance with the recent DHP in image super-resolution, meanwhile saves about 50% computation.

As an important computer vision task, image matching requires efficient and discriminative local descriptors. Most of the existing descriptors like SIFT and ORB are hand-crafted; therefore it is necessary to study more optimized descriptors through end-to-end learning. This paper proposes the compact binary descriptors learned with a lightweight Convolutional Neural Network (CNN), which is efficient for training and testing. Specifically, we propose a CNN with no larger than five layers for descriptor learning. The resulting descriptors, i.e., Compact Discriminative binary descriptors (CDbin) are optimized with four complementary loss functions, i.e., 1) triplet loss to ensure the discriminative power; 2) quantization loss to decrease the quantization error; 3) correlation loss to ensure the feature compactness; and 4) even-distribution loss to enrich the embedded information. The extensive experiments on two image patch datasets and three image retrieval datasets show that the CDbin exhibits competitive performance compared with the existing descriptors. For example, the 64-bit CDbin substantially outperforms the 256-bit ORB and 1024-bit SIFT on Hpatches dataset. Although generated by a shallow CNN, CDbin also outperforms several recent deep descriptors.

We present a scalable, hierarchical control system for the dynamic resource management of a distributed real-time embedded (DRE) system. This DRE is inspired by the DARPA adaptive and reflective middleware systems (ARMS) program. The goal of the control system is to simultaneously manage multiple resources and QoS concerns using a utility-driven approach for decision making and performance evaluation. At each level of the control hierarchy there are multiple local controllers which autonomously make decisions to optimize their local utility. The controllers in the hierarchy can use different, localized resource control algorithms and the system's user can tune the operations of the local controllers. We discuss how the selections of local control algorithms affect the behavior of the overall system. The control system is designed to be easily adaptable to other multi-tiered DRE systems.

In this paper we present a model-based approach for designing quality of service adaptive applications. We have developed a prototype distributed QoS modeling environment (DQME) that captures important elements of dynamic QoS adaptation at the model level. This modeling environment is designed independent of and can be integrated with, specific application domains to capture their QoS features and adaptation strategies. It combines the domain-specific modeling capability of the generic modeling environment with the QoS adaptation mechanisms of the quality objects middleware framework. DQME captures both the QoS and the functional concerns of distributed real-time embedded systems, and provides clear separation of these two. Integrated code-synthesis tools facilitate code generation and model refinement. We present a signal analyzer case study to demonstrate the use of the DQME modeling tool in real world applications.

Abstract We use a network model to choose retail prices in a grocery logistics system consisting of a warehouse (managed by a manufacturer) that supplies stores (managed by a retailer). We model retail prices as endogenous decisions that respond to store‐level cost and demand parameters. The retail prices are chosen to maximize the retailer's expected profits. The retail environment is captured by a three‐segment retail customer model where segments differ in their reservation price, size, and propensity to stockpile. The resulting network model chooses retail prices each period to optimize profits across the horizon. Retail price variation across time is profit‐maximizing because of the promotion sensitivity of the retail environment. We also model the impact of retail price variation on the manufacturer costs, warehouse inventory levels, and associated manufacturer expected profit. The suggested model is fit to a dataset consisting of retail prices and associated sales of canned soup. The model parameters as well as the residual variance are included in the decision model that generates the optimal retail prices that maximize the expected retail profit. We use model runs from the empirical dataset to show that (i) retail price variation can be profit‐maximizing for the retailer, (ii) retail market‐share constraints can decrease both manufacturer and retailer profits, and (iii) manufacturer price variation can be beneficial for the manufacturer and the retailer. The results show that manufacturer and retailer price variation can play a valuable role in improving logistics system performance in promotional retail environments. © 2001 John Wiley & Sons, Inc.