I Don't Know You, But I Can Catch You: Real-Time Defense against Diverse Adversarial Patches for Object Detectors

Deep neural networks (DNNs) have revolutionized the field of computer vision like object detection with their unparalleled performance. However, existing research has shown that DNNs are vulnerable to adversarial attacks. In the physical world, an adversary could exploit adversarial patches to implement a Hiding Attack (HA) which patches the target object to make it disappear from the detector, and an Appearing Attack (AA) which fools the detector into misclassifying the patch as a specific object. Recently, many defense methods for detectors have been proposed to mitigate the potential threats of adversarial patches. However, such methods still have limitations in generalization, robustness and efficiency. Most defenses are only effective against the HA, leaving the detector vulnerable to the AA. In this paper, we propose \textit{NutNet}, an innovative model for detecting adversarial patches, with high generalization, robustness and efficiency. With experiments for six detectors including YOLOv2-v4, SSD, Faster RCNN and DETR on both digital and physical domains, the results show that our proposed method can effectively defend against both the HA and AA, with only 0.4\% sacrifice of the clean performance. We compare NutNet with four baseline defense methods for detectors, and our method exhibits an average defense performance that is over 2.4 times and 4.7 times higher than existing approaches for HA and AA, respectively. In addition, NutNet only increases the inference time by 8\%, which can meet the real-time requirements of the detection systems. Demos of NutNet are available at: \url{https://sites.google.com/view/nutnet}.

DOEPatch: Dynamically Optimized Ensemble Model for Adversarial Patches Generation

Object detection is a fundamental task in various applications ranging from autonomous driving to intelligent security systems. However, recognition of a person can be hindered when their clothing is decorated with carefully designed graffiti patterns, leading to the failure of object detection. To achieve greater attack potential against unknown black-box models, adversarial patches capable of affecting the outputs of multiple-object detection models are required. While ensemble models have proven effective, current research in the field of object detection typically focuses on the simple fusion of the outputs of all models, with limited attention being given to developing general adversarial patches that can function effectively in the physical world. In this paper, we introduce the concept of energy and treat the adversarial patches generation process as an optimization of the adversarial patches to minimize the total energy of the ``person'' category. Additionally, by adopting adversarial training, we construct a dynamically optimized ensemble model. During training, the weight parameters of the attacked target models are adjusted to find the balance point at which the generated adversarial patches can effectively attack all target models. We carried out six sets of comparative experiments and tested our algorithm on five mainstream object detection models. The adversarial patches generated by our algorithm can reduce the recognition accuracy of YOLOv2 and YOLOv3 to 13.19\% and 29.20\%, respectively. In addition, we conducted experiments to test the effectiveness of T-shirts covered with our adversarial patches in the physical world and could achieve that people are not recognized by the object detection model. Finally, leveraging the Grad-CAM tool, we explored the attack mechanism of adversarial patches from an energetic perspective.

Area is all you need: repeatable elements make stronger adversarial attacks

Over the last decade, deep neural networks have achieved state of the art in computer vision tasks. These models, however, are susceptible to unusual inputs, known as adversarial examples, that cause them to misclassify or otherwise fail to detect objects. Here, we provide evidence that the increasing success of adversarial attacks is primarily due to increasing their size. We then demonstrate a method for generating the largest possible adversarial patch by building a adversarial pattern out of repeatable elements. This approach achieves a new state of the art in evading detection by YOLOv2 and YOLOv3. Finally, we present an experiment that fails to replicate the prior success of several attacks published in this field, and end with some comments on testing and reproducibility.

Improved Type III solar radio burst detection using congruent deep learning models

Solar flares are energetic events in the solar atmosphere that are often linked with solar radio bursts (SRBs). SRBs are observed at metric to decametric wavelengths and are classified into five spectral classes (Type I--V) based on their signature in dynamic spectra. The automatic detection and classification of SRBs is a challenge due to their heterogeneous form. Near-realtime detection and classification of SRBs has become a necessity in recent years due to large data rates generated by advanced radio telescopes such as the LOw Frequency ARray (LOFAR). In this study, we implement congruent deep learning models to automatically detect and classify Type III SRBs. We generated simulated Type III SRBs, which were comparable to Type IIIs seen in real observations, using a deep learning method known as Generative Adversarial Network (GAN). This simulated data was combined with observations from LOFAR to produce a training set that was used to train an object detection model known as YOLOv2 (You Only Look Once). Using this congruent deep learning model system, we can accurately detect Type III SRBs at a mean Average Precision (mAP) value of 77.71%.

YOWO-Plus: An Incremental Improvement

In this technical report, we would like to introduce our updates to YOWO, a real-time method for spatio-temporal action detection. We make a bunch of little design changes to make it better. For network structure, we use the same ones of official implemented YOWO, including 3D-ResNext-101 and YOLOv2, but we use a better pretrained weight of our reimplemented YOLOv2, which is better than the official YOLOv2. We also optimize the label assignment used in YOWO. To accurately detection action instances, we deploy GIoU loss for box regression. After our incremental improvement, YOWO achieves 84.9\% frame mAP and 50.5\% video mAP on the UCF101-24, significantly higher than the official YOWO. On the AVA, our optimized YOWO achieves 20.6\% frame mAP with 16 frames, also exceeding the official YOWO. With 32 frames, our YOWO achieves 21.6 frame mAP with 25 FPS on an RTX 3090 GPU. We name the optimized YOWO as YOWO-Plus. Moreover, we replace the 3D-ResNext-101 with the efficient 3D-ShuffleNet-v2 to design a lightweight action detector, YOWO-Nano. YOWO-Nano achieves 81.0 \% frame mAP and 49.7\% video frame mAP with over 90 FPS on the UCF101-24. It also achieves 18.4 \% frame mAP with about 90 FPS on the AVA. As far as we know, YOWO-Nano is the fastest state-of-the-art action detector. Our code is available on https://github.com/yjh0410/PyTorch_YOWO.

People detection and social distancing classification in smart cities for COVID-19 by using thermal images and deep learning algorithms

COVID-19 is a disease caused by severe respiratory syndrome coronavirus. It was identified in December 2019 in Wuhan, China. It has resulted in an ongoing pandemic that caused infected cases including some deaths. Coronavirus is primarily spread between people during close contact. Motivating to this notion, this research proposes an artificial intelligence system for social distancing classification of persons by using thermal images. By exploiting YOLOv2 (you look at once), a deep learning detection technique is developed for detecting and tracking people in indoor and outdoor scenarios. An algorithm is also implemented for measuring and classifying the distance between persons and automatically check if social distancing rules are respected or not. Hence, this work aims at minimizing the spread of the COVID-19 virus by evaluating if and how persons comply with social distancing rules. The proposed approach is applied to images acquired through thermal cameras, to establish a complete AI system for people tracking, social distancing classification, and body temperature monitoring. The training phase is done with two datasets captured from different thermal cameras. Ground Truth Labeler app is used for labeling the persons in the images. The achieved results show that the proposed method is suitable for the creation of a smart surveillance system in smart cities for people detection, social distancing classification, and body temperature analysis.

YOLO and Mask R-CNN for Vehicle Number Plate Identification

License plate scanners have grown in popularity in parking lots during the past few years. In order to quickly identify license plates, traditional plate recognition devices used in parking lots employ a fixed source of light and shooting angles. For skewed angles, such as license plate images taken with ultra-wide angle or fisheye lenses, deformation of the license plate recognition plate can also be quite severe, impairing the ability of standard license plate recognition systems to identify the plate. Mask RCNN gadget that may be utilised for oblique pictures and various shooting angles. The results of the experiments show that the suggested design will be capable of classifying license plates with bevel angles larger than 0/60. Character recognition using the suggested Mask R-CNN approach has advanced significantly as well. The proposed Mask R-CNN method has also achieved significant progress in character recognition, which is tilted more than 45 degrees as compared to the strategy of employing the YOLOv2 model. Experiment results also suggest that the methodology presented in the open data plate collecting is better than other techniques (known as the AOLP dataset).

Hybrid Classifiers for Spatio-temporal Real-time Abnormal Behaviors Detection, Tracking, and Recognition in Massive Hajj Crowds

Individual abnormal behaviors vary depending on crowd sizes, contexts, and scenes. Challenges such as partial occlusions, blurring, large-number abnormal behavior, and camera viewing occur in large-scale crowds when detecting, tracking, and recognizing individuals with abnormal behaviors. In this paper, our contribution is twofold. First, we introduce an annotated and labeled large-scale crowd abnormal behaviors Hajj dataset (HAJJv2). Second, we propose two methods of hybrid Convolutional Neural Networks (CNNs) and Random Forests (RFs) to detect and recognize Spatio-temporal abnormal behaviors in small and large-scales crowd videos. In small-scale crowd videos, a ResNet-50 pre-trained CNN model is fine-tuned to verify whether every frame is normal or abnormal in the spatial domain. If anomalous behaviors are observed, a motion-based individuals detection method based on the magnitudes and orientations of Horn-Schunck optical flow is used to locate and track individuals with abnormal behaviors. A Kalman filter is employed in large-scale crowd videos to predict and track the detected individuals in the subsequent frames. Then, means, variances, and standard deviations statistical features are computed and fed to the RF to classify individuals with abnormal behaviors in the temporal domain. In large-scale crowds, we fine-tune the ResNet-50 model using YOLOv2 object detection technique to detect individuals with abnormal behaviors in the spatial domain.

A Real Time 1280x720 Object Detection Chip With 585MB/s Memory Traffic

Memory bandwidth has become the real-time bottleneck of current deep learning accelerators (DLA), particularly for high definition (HD) object detection. Under resource constraints, this paper proposes a low memory traffic DLA chip with joint hardware and software optimization. To maximize hardware utilization under memory bandwidth, we morph and fuse the object detection model into a group fusion-ready model to reduce intermediate data access. This reduces the YOLOv2's feature memory traffic from 2.9 GB/s to 0.15 GB/s. To support group fusion, our previous DLA based hardware employes a unified buffer with write-masking for simple layer-by-layer processing in a fusion group. When compared to our previous DLA with the same PE numbers, the chip implemented in a TSMC 40nm process supports 1280x720@30FPS object detection and consumes 7.9X less external DRAM access energy, from 2607 mJ to 327.6 mJ.

Detecting Cloud-Based Phishing Attacks by Combining Deep Learning Models

Web-based phishing attacks nowadays exploit popular cloud web hosting services and apps such as Google Sites and Typeform for hosting their attacks. Since these attacks originate from reputable domains and IP addresses of the cloud services, traditional phishing detection methods such as IP reputation monitoring and blacklisting are not very effective. Here we investigate the effectiveness of deep learning models in detecting this class of cloud-based phishing attacks. Specifically, we evaluate deep learning models for three phishing detection methods--LSTM model for URL analysis, YOLOv2 model for logo analysis, and triplet network model for visual similarity analysis. We train the models using well-known datasets and test their performance on phishing attacks in the wild. Our results qualitatively explain why the models succeed or fail. Furthermore, our results highlight how combining results from the individual models can improve the effectiveness of detecting cloud-based phishing attacks.