In this paper, we present InSeGAN, an unsupervised 3D generative adversarial network (GAN) for segmenting (nearly) identical instances of rigid objects in depth images. Using an analysis-by-synthesis approach, we design a novel GAN architecture to synthesize a multiple-instance depth image with independent control over each instance. InSeGAN takes in a set of code vectors (e.g., random noise vectors), each encoding the 3D pose of an object that is represented by a learned implicit object template. The generator has two distinct modules. The first module, the instance feature generator, uses each encoded pose to transform the implicit template into a feature map representation of each object instance. The second module, the depth image renderer, aggregates all of the single-instance feature maps output by the first module and generates a multiple-instance depth image. A discriminator distinguishes the generated multiple-instance depth images from the distribution of true depth images. To use our model for instance segmentation, we propose an instance pose encoder that learns to take in a generated depth image and reproduce the pose code vectors for all of the object instances. To evaluate our approach, we introduce a new synthetic dataset, "Insta-10", consisting of 100,000 depth images, each with 5 instances of an object from one of 10 classes. Our experiments on Insta-10, as well as on real-world noisy depth images, show that InSeGAN achieves state-of-the-art performance, often outperforming prior methods by large margins.
This report summarizes the results of Learning to Understand Aerial Images (LUAI) 2021 challenge held on ICCV 2021, which focuses on object detection and semantic segmentation in aerial images. Using DOTA-v2.0 and GID-15 datasets, this challenge proposes three tasks for oriented object detection, horizontal object detection, and semantic segmentation of common categories in aerial images. This challenge received a total of 146 registrations on the three tasks. Through the challenge, we hope to draw attention from a wide range of communities and call for more efforts on the problems of learning to understand aerial images.
Video privacy leakage is becoming an increasingly severe public problem, especially in cloud-based video surveillance systems. It leads to the new need for secure cloud-based video applications, where the video is encrypted for privacy protection. Despite some methods that have been proposed for encrypted video moving object detection and tracking, none has robust performance against complex and dynamic scenes. In this paper, we propose an efficient and robust privacy-preserving motion detection and multiple object tracking scheme for encrypted surveillance video bitstreams. By analyzing the properties of the video codec and format-compliant encryption schemes, we propose a new compressed-domain feature to capture motion information in complex surveillance scenarios. Based on this feature, we design an adaptive clustering algorithm for moving object segmentation with an accuracy of 4x4 pixels. We then propose a multiple object tracking scheme that uses Kalman filter estimation and adaptive measurement refinement. The proposed scheme does not require video decryption or full decompression and has a very low computation load. The experimental results demonstrate that our scheme achieves the best detection and tracking performance compared with existing works in the encrypted and compressed domain. Our scheme can be effectively used in complex surveillance scenarios with different challenges, such as camera movement/jitter, dynamic background, and shadows.
The goal of few-shot fine-grained image classification is to recognize rarely seen fine-grained objects in the query set, given only a few samples of this class in the support set. Previous works focus on learning discriminative image features from a limited number of training samples for distinguishing various fine-grained classes, but ignore one important fact that spatial alignment of the discriminative semantic features between the query image with arbitrary changes and the support image, is also critical for computing the semantic similarity between each support-query pair. In this work, we propose an object-aware long-short-range spatial alignment approach, which is composed of a foreground object feature enhancement (FOE) module, a long-range semantic correspondence (LSC) module and a short-range spatial manipulation (SSM) module. The FOE is developed to weaken background disturbance and encourage higher foreground object response. To address the problem of long-range object feature misalignment between support-query image pairs, the LSC is proposed to learn the transferable long-range semantic correspondence by a designed feature similarity metric. Further, the SSM module is developed to refine the transformed support feature after the long-range step to align short-range misaligned features (or local details) with the query features. Extensive experiments have been conducted on four benchmark datasets, and the results show superior performance over most state-of-the-art methods under both 1-shot and 5-shot classification scenarios.
As one of the basic tasks of computer vision, object detection has been widely used in many intelligent applications. However, object detection algorithms are usually heavyweight in computation, hindering their implementations on resource-constrained edge devices. Current edge-cloud collaboration methods, such as CNN partition over Edge-cloud devices, are not suitable for object detection since the huge data size of the intermediate results will introduce extravagant communication costs. To address this challenge, we propose a small-big model framework that deploys a big model in the cloud and a small model on the edge devices. Upon receiving data, the edge device operates a difficult-case discriminator to classify the images into easy cases and difficult cases according to the specific semantics of the images. The easy cases will be processed locally at the edge, and the difficult cases will be uploaded to the cloud. Experimental results on the VOC, COCO, HELMET datasets using two different object detection algorithms demonstrate that the small-big model system can detect 94.01%-97.84% of objects with only about 50% images uploaded to the cloud when using SSD. In addition, the small-big model averagely reaches 91.22%- 92.52% end-to-end mAP of the scheme that uploading all images to the cloud.
This work tackles the unsupervised cross-domain object detection problem which aims to generalize a pre-trained object detector to a new target domain without labels. We propose an uncertainty-aware model adaptation method, which is based on two motivations: 1) the estimation and exploitation of model uncertainty in a new domain is critical for reliable domain adaptation; and 2) the joint alignment of distributions for inputs (feature alignment) and outputs (self-training) is needed. To this end, we compose a Bayesian CNN-based framework for uncertainty estimation in object detection, and propose an algorithm for generation of uncertainty-aware pseudo-labels. We also devise a scheme for joint feature alignment and self-training of the object detection model with uncertainty-aware pseudo-labels. Experiments on multiple cross-domain object detection benchmarks show that our proposed method achieves state-of-the-art performance.
Automatic detection of prohibited items within complex and cluttered X-ray security imagery is essential to maintaining transport security, where prior work on automatic prohibited item detection focus primarily on pseudo-colour (rgb}) X-ray imagery. In this work we study the impact of variant X-ray imagery, i.e., X-ray energy response (high, low}) and effective-z compared to rgb, via the use of deep Convolutional Neural Networks (CNN) for the joint object detection and segmentation task posed within X-ray baggage security screening. We evaluate state-of-the-art CNN architectures (Mask R-CNN, YOLACT, CARAFE and Cascade Mask R-CNN) to explore the transferability of models trained with such 'raw' variant imagery between the varying X-ray security scanners that exhibits differing imaging geometries, image resolutions and material colour profiles. Overall, we observe maximal detection performance using CARAFE, attributable to training using combination of rgb, high, low, and effective-z X-ray imagery, obtaining 0.7 mean Average Precision (mAP) for a six class object detection problem. Our results also exhibit a remarkable degree of generalisation capability in terms of cross-scanner transferability (AP: 0.835/0.611) for a one class object detection problem by combining rgb, high, low, and effective-z imagery.
This study investigates an application of a new probabilistic interpretation of a softmax output to Open-Set Recognition (OSR). Softmax is a mechanism wildly used in classification and object recognition. However, a softmax mechanism forces a model to operate under a closed-set paradigm, i.e., to predict an object class out of a set of pre-defined labels. This characteristic contributes to efficacy in classification, but poses a risk of non-sense prediction in object recognition. Object recognition is often operated under a dynamic and diverse condition. A foreign object -- an object of any unprepared class -- can be encountered at any time. OSR is intended to address an issue of identifying a foreign object in object recognition. Based on Bayes theorem and the emphasis of conditioning on the context, softmax inference has been re-interpreted. This re-interpretation has led to a new approach to OSR, called Latent Cognizance (LC). Our investigation employs various scenarios, using Imagenet 2012 dataset as well as fooling and open-set images. The findings support LC hypothesis and show its effectiveness on OSR.
Blood cell detection in microscopic images is an essential branch of medical image processing research. Since disease detection based on manual checking of blood cells is time-consuming and full of errors, testing of blood cells using object detectors with Deep Convolutional Neural Network can be regarded as a feasible solution. In this work, an object detector based on YOLOF has been proposed to detect blood cell objects such as red blood cells, white blood cells and platelets. This object detector is called TE-YOLOF, Tiny and Efficient YOLOF, and it is a One-Stage detector using dilated encoder to extract information from single-level feature maps. For increasing efficiency and flexibility, the EfficientNet Convolutional Neural Network is utilized as the backbone for the proposed object detector. Furthermore, the Depthwise Separable Convolution is applied to enhance the performance and minimize the parameters of the network. In addition, the Mish activation function is employed to increase the precision. Extensive experiments on the BCCD dataset prove the effectiveness of the proposed model, which is more efficient than other existing studies for blood cell detection.
Object detection aims to localize and classify the objects in a given image, and these two tasks are sensitive to different object regions. Therefore, some locations predict high-quality bounding boxes but low classification scores, and some locations are quite the opposite. A misalignment exists between the two tasks, and their features are spatially entangled. In order to solve the misalignment problem, we propose a plug-in Spatial-disentangled and Task-aligned operator (SALT). By predicting two task-aware point sets that are located in each task's sensitive regions, SALT can reassign features from those regions and align them to the corresponding anchor point. Therefore, features for the two tasks are spatially aligned and disentangled. To minimize the difference between the two regression stages, we propose a Self-distillation regression (SDR) loss that can transfer knowledge from the refined regression results to the coarse regression results. On the basis of SALT and SDR loss, we propose SALT-Net, which explicitly exploits task-aligned point-set features for accurate detection results. Extensive experiments on the MS-COCO dataset show that our proposed methods can consistently boost different state-of-the-art dense detectors by $\sim$2 AP. Notably, SALT-Net with Res2Net-101-DCN backbone achieves 53.8 AP on the MS-COCO test-dev.
Deep neural networks, especially convolutional deep neural networks, are state-of-the-art methods to classify, segment or even generate images, movies, or sounds. However, these methods lack of a good semantic understanding of what happens internally. The question, why a COVID-19 detector has classified a stack of lung-ct images as positive, is sometimes more interesting than the overall specificity and sensitivity. Especially when human domain expert knowledge disagrees with the given output. This way, human domain experts could also be advised to reconsider their choice, regarding the information pointed out by the system. In addition, the deep learning model can be controlled, and a present dataset bias can be found. Currently, most explainable AI methods in the computer vision domain are purely used on image classification, where the images are ordinary images in the visible spectrum. As a result, there is no comparison on how the methods behave with multimodal image data, as well as most methods have not been investigated on how they behave when used for object detection. This work tries to close the gaps. Firstly, investigating three saliency map generator methods on how their maps differ across the different spectra. This is achieved via accurate and systematic training. Secondly, we examine how they behave when used for object detection. As a practical problem, we chose object detection in the infrared and visual spectrum for autonomous driving. The dataset used in this work is the Multispectral Object Detection Dataset, where each scene is available in the FIR, MIR and NIR as well as visual spectrum. The results show that there are differences between the infrared and visual activation maps. Further, an advanced training with both, the infrared and visual data not only improves the network's output, it also leads to more focused spots in the saliency maps.
Deep neural networks (DNNs) have become essential for processing the vast amounts of aerial imagery collected using earth-observing satellite platforms. However, DNNs are vulnerable towards adversarial examples, and it is expected that this weakness also plagues DNNs for aerial imagery. In this work, we demonstrate one of the first efforts on physical adversarial attacks on aerial imagery, whereby adversarial patches were optimised, fabricated and installed on or near target objects (cars) to significantly reduce the efficacy of an object detector applied on overhead images. Physical adversarial attacks on aerial images, particularly those captured from satellite platforms, are challenged by atmospheric factors (lighting, weather, seasons) and the distance between the observer and target. To investigate the effects of these challenges, we devised novel experiments and metrics to evaluate the efficacy of physical adversarial attacks against object detectors in aerial scenes. Our results indicate the palpable threat posed by physical adversarial attacks towards DNNs for processing satellite imagery.
With the development of underwater object grabbing technology, underwater object recognition and segmentation of high accuracy has become a challenge. The existing underwater object detection technology can only give the general position of an object, unable to give more detailed information such as the outline of the object, which seriously affects the grabbing efficiency. To address this problem, we label and establish the first underwater semantic segmentation dataset of real scene(DUT-USEG:DUT Underwater Segmentation Dataset). The DUT- USEG dataset includes 6617 images, 1487 of which have semantic segmentation and instance segmentation annotations, and the remaining 5130 images have object detection box annotations. Based on this dataset, we propose a semi-supervised underwater semantic segmentation network focusing on the boundaries(US-Net: Underwater Segmentation Network). By designing a pseudo label generator and a boundary detection subnetwork, this network realizes the fine learning of boundaries between underwater objects and background, and improves the segmentation effect of boundary areas. Experiments show that the proposed method improves by 6.7% in three categories of holothurian, echinus, starfish in DUT-USEG dataset, and achieves state-of-the-art results. The DUT- USEG dataset will be released at https://github.com/baxiyi/DUT-USEG.
In this paper, given a small bag of images, each containing a common but latent predicate, we are interested in localizing visual subject-object pairs connected via the common predicate in each of the images. We refer to this novel problem as visual relationship co-localization or VRC as an abbreviation. VRC is a challenging task, even more so than the well-studied object co-localization task. This becomes further challenging when using just a few images, the model has to learn to co-localize visual subject-object pairs connected via unseen predicates. To solve VRC, we propose an optimization framework to select a common visual relationship in each image of the bag. The goal of the optimization framework is to find the optimal solution by learning visual relationship similarity across images in a few-shot setting. To obtain robust visual relationship representation, we utilize a simple yet effective technique that learns relationship embedding as a translation vector from visual subject to visual object in a shared space. Further, to learn visual relationship similarity, we utilize a proven meta-learning technique commonly used for few-shot classification tasks. Finally, to tackle the combinatorial complexity challenge arising from an exponential number of feasible solutions, we use a greedy approximation inference algorithm that selects approximately the best solution. We extensively evaluate our proposed framework on variations of bag sizes obtained from two challenging public datasets, namely VrR-VG and VG-150, and achieve impressive visual co-localization performance.
Object detection on drone-captured scenarios is a recent popular task. As drones always navigate in different altitudes, the object scale varies violently, which burdens the optimization of networks. Moreover, high-speed and low-altitude flight bring in the motion blur on the densely packed objects, which leads to great challenge of object distinction. To solve the two issues mentioned above, we propose TPH-YOLOv5. Based on YOLOv5, we add one more prediction head to detect different-scale objects. Then we replace the original prediction heads with Transformer Prediction Heads (TPH) to explore the prediction potential with self-attention mechanism. We also integrate convolutional block attention model (CBAM) to find attention region on scenarios with dense objects. To achieve more improvement of our proposed TPH-YOLOv5, we provide bags of useful strategies such as data augmentation, multiscale testing, multi-model integration and utilizing extra classifier. Extensive experiments on dataset VisDrone2021 show that TPH-YOLOv5 have good performance with impressive interpretability on drone-captured scenarios. On DET-test-challenge dataset, the AP result of TPH-YOLOv5 are 39.18%, which is better than previous SOTA method (DPNetV3) by 1.81%. On VisDrone Challenge 2021, TPHYOLOv5 wins 5th place and achieves well-matched results with 1st place model (AP 39.43%). Compared to baseline model (YOLOv5), TPH-YOLOv5 improves about 7%, which is encouraging and competitive.
We explore object detection with two attributes: color and material. The task aims to simultaneously detect objects and infer their color and material. A straight-forward approach is to add attribute heads at the very end of a usual object detection pipeline. However, we observe that the two goals are in conflict: Object detection should be attribute-independent and attributes be largely object-independent. Features computed by a standard detection network entangle the category and attribute features; we disentangle them by the use of a two-stream model where the category and attribute features are computed independently but the classification heads share Regions of Interest (RoIs). Compared with a traditional single-stream model, our model shows significant improvements over VG-20, a subset of Visual Genome, on both supervised and attribute transfer tasks.
Although modern object detection and classification models achieve high accuracy, these are typically constrained in advance on a fixed train set and are therefore not flexible to deal with novel, unseen object categories. Moreover, these models most often operate on a single frame, which may yield incorrect classifications in case of ambiguous viewpoints. In this paper, we propose an active inference agent that actively gathers evidence for object classifications, and can learn novel object categories over time. Drawing inspiration from the human brain, we build object-centric generative models composed of two information streams, a what- and a where-stream. The what-stream predicts whether the observed object belongs to a specific category, while the where-stream is responsible for representing the object in its internal 3D reference frame. We show that our agent (i) is able to learn representations for many object categories in an unsupervised way, (ii) achieves state-of-the-art classification accuracies, actively resolving ambiguity when required and (iii) identifies novel object categories. Furthermore, we validate our system in an end-to-end fashion where the agent is able to search for an object at a given pose from a pixel-based rendering. We believe that this is a first step towards building modular, intelligent systems that can be used for a wide range of tasks involving three dimensional objects.
One of the most difficult jobs in remote sensing is dealing with traffic bottlenecks at airports. This fact has been confirmed by several studies attempting to resolve this issue. Among a wide range of approaches employed the most successful methods have been based on airplane object recognition using satellite images as datasets for deep learning models. Airplane object identification is not a viable method for resolving traffic congestion. There are several types of airplanes at the airport each with its own set of requirements and specifications.Utilizing satellite pictures will require the use of complex equipment which is a financial burden. In this work a universal low-cost and efficient solution for airport traffic congestion is proposed. Drone-captured aerial pictures are used to train and assess a Mask Region Convolution Neural Network model. This model can detect the presence of any aircraft in a photograph and pinpoint its location. It also includes mask estimations to properly identify each detected airplane type based on the estimated surface area and cabin length which are crucial variables in distinguishing planes. The study is conducted using Microsoft Common Object in Context COCO metrics average accuracies and a confusion matrix all of which demonstrate the approach's potential in providing valuable aid for dealing with traffic congestion at airports.
This paper presents INVIGORATE, a robot system that interacts with human through natural language and grasps a specified object in clutter. The objects may occlude, obstruct, or even stack on top of one another. INVIGORATE embodies several challenges: (i) infer the target object among other occluding objects, from input language expressions and RGB images, (ii) infer object blocking relationships (OBRs) from the images, and (iii) synthesize a multi-step plan to ask questions that disambiguate the target object and to grasp it successfully. We train separate neural networks for object detection, for visual grounding, for question generation, and for OBR detection and grasping. They allow for unrestricted object categories and language expressions, subject to the training datasets. However, errors in visual perception and ambiguity in human languages are inevitable and negatively impact the robot's performance. To overcome these uncertainties, we build a partially observable Markov decision process (POMDP) that integrates the learned neural network modules. Through approximate POMDP planning, the robot tracks the history of observations and asks disambiguation questions in order to achieve a near-optimal sequence of actions that identify and grasp the target object. INVIGORATE combines the benefits of model-based POMDP planning and data-driven deep learning. Preliminary experiments with INVIGORATE on a Fetch robot show significant benefits of this integrated approach to object grasping in clutter with natural language interactions. A demonstration video is available at https://youtu.be/zYakh80SGcU.
Information-driven control can be used to develop intelligent sensors that can optimize their measurement value based on environmental feedback. In object tracking applications, sensor actions are chosen based on the expected reduction in uncertainty also known as information gain. Random finite set (RFS) theory provides a formalism for quantifying and estimating information gain in multi-object tracking problems. However, estimating information gain in these applications remains computationally challenging. This paper presents a new tractable approximation of the RFS expected information gain applicable to sensor control for multi-object search and tracking. Unlike existing RFS approaches, the information gain approximation presented in this paper considers the contributions of non-ideal noisy measurements, missed detections, false alarms, and object appearance/disappearance. The effectiveness of the information-driven sensor control is demonstrated through two multi-vehicle search-while-tracking experiments using real video data from remote terrestrial and satellite sensors.