Image colorization is inherently an ill-posed problem with multi-modal uncertainty. Previous methods leverage the deep neural network to map input grayscale images to plausible color outputs directly. Although these learning-based methods have shown impressive performance, they usually fail on the input images that contain multiple objects. The leading cause is that existing models perform learning and colorization on the entire image. In the absence of a clear figure-ground separation, these models cannot effectively locate and learn meaningful object-level semantics. In this paper, we propose a method for achieving instance-aware colorization. Our network architecture leverages an off-the-shelf object detector to obtain cropped object images and uses an instance colorization network to extract object-level features. We use a similar network to extract the full-image features and apply a fusion module to full object-level and image-level features to predict the final colors. Both colorization networks and fusion modules are learned from a large-scale dataset. Experimental results show that our work outperforms existing methods on different quality metrics and achieves state-of-the-art performance on image colorization.
Deep Convolutional Neural Networks (CNNs) have been repeatedly proven to perform well on image classification tasks. Object detection methods, however, are still in need of significant improvements. In this paper, we propose a new framework called Ventral-Dorsal Networks (VDNets) which is inspired by the structure of the human visual system. Roughly, the visual input signal is analyzed along two separate neural streams, one in the temporal lobe and the other in the parietal lobe. The coarse functional distinction between these streams is between object recognition -- the "what" of the signal -- and extracting location related information -- the "where" of the signal. The ventral pathway from primary visual cortex, entering the temporal lobe, is dominated by "what" information, while the dorsal pathway, into the parietal lobe, is dominated by "where" information. Inspired by this structure, we propose the integration of a "Ventral Network" and a "Dorsal Network", which are complementary. Information about object identity can guide localization, and location information can guide attention to relevant image regions, improving object recognition. This new dual network framework sharpens the focus of object detection. Our experimental results reveal that the proposed method outperforms state-of-the-art object detection approaches on PASCAL VOC 2007 by 8% (mAP) and PASCAL VOC 2012 by 3% (mAP). Moreover, a comparison of techniques on Yearbook images displays substantial qualitative and quantitative benefits of VDNet.
Visual identification of gunmen in a crowd is a challenging problem, that requires resolving the association of a person with an object (firearm). We present a novel approach to address this problem, by defining human-object interaction (and non-interaction) bounding boxes. In a given image, human and firearms are separately detected. Each detected human is paired with each detected firearm, allowing us to create a paired bounding box that contains both object and the human. A network is trained to classify these paired-bounding-boxes into human carrying the identified firearm or not. Extensive experiments were performed to evaluate effectiveness of the algorithm, including exploiting full pose of the human, hand key-points, and their association with the firearm. The knowledge of spatially localized features is key to success of our method by using multi-size proposals with adaptive average pooling. We have also extended a previously firearm detection dataset, by adding more images and tagging in extended dataset the human-firearm pairs (including bounding boxes for firearms and gunmen). The experimental results ($AP_{hold} = 78.5$) demonstrate effectiveness of the proposed method.
Adversarial examples are data points misclassified by neural networks. Originally, adversarial examples were limited to adding small perturbations to a given image. Recent work introduced the generalized concept of unrestricted adversarial examples, without limits on the added perturbations. In this paper, we introduce a new category of attacks that create unrestricted adversarial examples for object detection. Our key idea is to generate adversarial objects that are unrelated to the classes identified by the target object detector. Different from previous attacks, we use off-the-shelf Generative Adversarial Networks (GAN), without requiring any further training or modification. Our method consists of searching over the latent normal space of the GAN for adversarial objects that are wrongly identified by the target object detector. We evaluate this method on the commonly used Faster R-CNN ResNet-101, Inception v2 and SSD Mobilenet v1 object detectors using logo generative iWGAN-LC and SNGAN trained on CIFAR-10. The empirical results show that the generated adversarial objects are indistinguishable from non-adversarial objects generated by the GANs, transferable between the object detectors and robust in the physical world. This is the first work to study unrestricted false positive adversarial examples for object detection.
We propose a new deep convolutional neural network framework that uses object location knowledge implicit in network connection weights to guide selective attention in object detection tasks. Our approach is called What-Where Nets (WW-Nets), and it is inspired by the structure of human visual pathways. In the brain, vision incorporates two separate streams, one in the temporal lobe and the other in the parietal lobe, called the ventral stream and the dorsal stream, respectively. The ventral pathway from primary visual cortex is dominated by "what" information, while the dorsal pathway is dominated by "where" information. Inspired by this structure, we have proposed an object detection framework involving the integration of a "What Network" and a "Where Network". The aim of the What Network is to provide selective attention to the relevant parts of the input image. The Where Network uses this information to locate and classify objects of interest. In this paper, we compare this approach to state-of-the-art algorithms on the PASCAL VOC 2007 and 2012 and COCO object detection challenge datasets. Also, we compare out approach to human "ground-truth" attention. We report the results of an eye-tracking experiment on human subjects using images from PASCAL VOC 2007, and we demonstrate interesting relationships between human overt attention and information processing in our WW-Nets. Finally, we provide evidence that our proposed method performs favorably in comparison to other object detection approaches, often by a large margin. The code and the eye-tracking ground-truth dataset can be found at: https://github.com/mkebrahimpour.
In the domain of autonomous driving, deep learning has substantially improved the 3D object detection accuracy for LiDAR and stereo camera data alike. While deep networks are great at generalization, they are also notorious to over-fit to all kinds of spurious artifacts, such as brightness, car sizes and models, that may appear consistently throughout the data. In fact, most datasets for autonomous driving are collected within a narrow subset of cities within one country, typically under similar weather conditions. In this paper we consider the task of adapting 3D object detectors from one dataset to another. We observe that naively, this appears to be a very challenging task, resulting in drastic drops in accuracy levels. We provide extensive experiments to investigate the true adaptation challenges and arrive at a surprising conclusion: the primary adaptation hurdle to overcome are differences in car sizes across geographic areas. A simple correction based on the average car size yields a strong correction of the adaptation gap. Our proposed method is simple and easily incorporated into most 3D object detection frameworks. It provides a first baseline for 3D object detection adaptation across countries, and gives hope that the underlying problem may be more within grasp than one may have hoped to believe. Our code is available at https://github.com/cxy1997/3D_adapt_auto_driving.
We describe an Object Oriented Model for building Expert Systems. This model and the detection of similarities allow to implement reasoning modes as induction, deduction and simulation. We specially focus on similarity and its use in induction. We propose original algorithms which deal with total and partial structural similitude of objects to facilitate knowledge acquisition.
Training with more data has always been the most stable and effective way of improving performance in deep learning era. As the largest object detection dataset so far, Open Images brings great opportunities and challenges for object detection in general and sophisticated scenarios. However, owing to its semi-automatic collecting and labeling pipeline to deal with the huge data scale, Open Images dataset suffers from label-related problems that objects may explicitly or implicitly have multiple labels and the label distribution is extremely imbalanced. In this work, we quantitatively analyze these label problems and provide a simple but effective solution. We design a concurrent softmax to handle the multi-label problems in object detection and propose a soft-sampling methods with hybrid training scheduler to deal with the label imbalance. Overall, our method yields a dramatic improvement of 3.34 points, leading to the best single model with 60.90 mAP on the public object detection test set of Open Images. And our ensembling result achieves 67.17 mAP, which is 4.29 points higher than the best result of Open Images public test 2018.
We are currently designing an object oriented model which describes static and dynamical knowledge in diff{\'e}rent domains. It provides a twin conceptual level. The internal level proposes: the object structure composed of sub-objects hierarchy, structure evolution with dynamical functions, same type objects comparison with evaluation functions. It uses multiple upward inheritance from sub-objects properties to the Object. The external level describes: object environment, it enforces object types and uses external simple inheritance from the type to the sub-types.
The ability to recognize objects despite there being differences in appearance, known as Core Object Recognition, forms a critical part of human perception. While it is understood that the brain accomplishes Core Object Recognition through feedforward, hierarchical computations through the visual stream, the underlying algorithms that allow for invariant representations to form downstream is still not well understood. (DiCarlo et al., 2012) Various computational perceptual models have been built to attempt and tackle the object identification task in an artificial perceptual setting. Artificial Neural Networks, computational graphs consisting of weighted edges and mathematical operations at vertices, are loosely inspired by neural networks in the brain and have proven effective at various visual perceptual tasks, including object characterization and identification. (Pinto et al., 2008) (DiCarlo et al., 2012) For many data analysis tasks, learning representations where each dimension is statistically independent and thus disentangled from the others is useful. If the underlying generative factors of the data are also statistically independent, Bayesian inference of latent variables can form disentangled representations. This thesis constitutes a research project exploring a generalization of the Variational Autoencoder (VAE), $\beta$-VAE, that aims to learn disentangled representations using variational inference. $\beta$-VAE incorporates the hyperparameter $\beta$, and enforces conditional independence of its bottleneck neurons, which is in general not compatible with the statistical independence of latent variables. This text examines this architecture, and provides analytical and numerical arguments, with the goal of demonstrating that this incompatibility leads to a non-monotonic inference performance in $\beta$-VAE with a finite optimal $\beta$.
Instance segmentation is one of the actively studied research topics in computer vision in which many objects of interest should be separated individually. While many feed-forward networks produce high-quality segmentation on different types of images, their results often suffer from topological errors (merging or splitting) for segmentation of many objects, requiring post-processing. Existing iterative methods, on the other hand, extract a single object at a time using discriminative knowledge-based properties (shapes, boundaries, etc.) without relying on post-processing, but they do not scale well. To exploit the advantages of conventional single-object-per-step segmentation methods without impairing the scalability, we propose a novel iterative deep reinforcement learning agent that learns how to differentiate multiple objects in parallel. Our reward function for the trainable agent is designed to favor grouping pixels belonging to the same object using a graph coloring algorithm. We demonstrate that the proposed method can efficiently perform instance segmentation of many objects without heavy post-processing.
A structured query can capture the complexity of object interactions (e.g. 'woman rides motorcycle') unlike single objects (e.g. 'woman' or 'motorcycle'). Retrieval using structured queries therefore is much more useful than single object retrieval, but a much more challenging problem. In this paper we present a method which uses scene graph embeddings as the basis for an approach to image retrieval. We examine how visual relationships, derived from scene graphs, can be used as structured queries. The visual relationships are directed subgraphs of the scene graph with a subject and object as nodes connected by a predicate relationship. Notably, we are able to achieve high recall even on low to medium frequency objects found in the long-tailed COCO-Stuff dataset, and find that adding a visual relationship-inspired loss boosts our recall by 10% in the best case.
This paper introduces Action Image, a new grasp proposal representation that allows learning an end-to-end deep-grasping policy. Our model achieves $84\%$ grasp success on $172$ real world objects while being trained only in simulation on $48$ objects with just naive domain randomization. Similar to computer vision problems, such as object detection, Action Image builds on the idea that object features are invariant to translation in image space. Therefore, grasp quality is invariant when evaluating the object-gripper relationship; a successful grasp for an object depends on its local context, but is independent of the surrounding environment. Action Image represents a grasp proposal as an image and uses a deep convolutional network to infer grasp quality. We show that by using an Action Image representation, trained networks are able to extract local, salient features of grasping tasks that generalize across different objects and environments. We show that this representation works on a variety of inputs, including color images (RGB), depth images (D), and combined color-depth (RGB-D). Our experimental results demonstrate that networks utilizing an Action Image representation exhibit strong domain transfer between training on simulated data and inference on real-world sensor streams. Finally, our experiments show that a network trained with Action Image improves grasp success ($84\%$ vs. $53\%$) over a baseline model with the same structure, but using actions encoded as vectors.
Deep learning-based detectors usually produce a redundant set of object bounding boxes including many duplicate detections of the same object. These boxes are then filtered using non-maximum suppression (NMS) in order to select exactly one bounding box per object of interest. This greedy scheme is simple and provides sufficient accuracy for isolated objects but often fails in crowded environments, since one needs to both preserve boxes for different objects and suppress duplicate detections. In this work we develop an alternative iterative scheme, where a new subset of objects is detected at each iteration. Detected boxes from the previous iterations are passed to the network at the following iterations to ensure that the same object would not be detected twice. This iterative scheme can be applied to both one-stage and two-stage object detectors with just minor modifications of the training and inference procedures. We perform extensive experiments with two different baseline detectors on four datasets and show significant improvement over the baseline, leading to state-of-the-art performance on CrowdHuman and WiderPerson datasets. The source code and the trained models are available at https://github.com/saic-vul/iterdet.
Deep neural network based learning approaches is widely utilized for image classification or object detection based problems with remarkable outcomes. Realtime Object state estimation of objects can be used to track and estimate the features that the object of the current frame possesses without causing any significant delay and misclassification. A system that can detect the features of such objects in the present state from camera images can be used to enhance the application of Augmented Reality for improving user experience and delivering information in a much perceptual way. The focus behind this paper is to determine the most suitable model to create a low-latency assistance AR to aid users by providing them nutritional information about the food that they consume in order to promote healthier life choices. Hence the dataset has been collected and acquired in such a manner, and we conduct various tests in order to identify the most suitable DNN in terms of performance and complexity and establish a system that renders such information realtime to the user.
Existing anchor-based and anchor-free object detectors in multi-stage or one-stage pipelines have achieved very promising detection performance. However, they still encounter the design difficulty in hand-crafted 2D anchor definition and the learning complexity in 1D direct location regression. To tackle these issues, in this paper, we propose a novel detector coined as ScopeNet, which models anchors of each location as a mutually dependent relationship. This approach quantizes the prediction space and employs a coarse-to-fine strategy for localization. It achieves superior flexibility as in the regression based anchor-free methods, while produces more precise prediction. Besides, an inherit anchor selection score is learned to indicate the localization quality of the detection result, and we propose to better represent the confidence of a detection box by combining the category-classification score and the anchor-selection score. With our concise and effective design, the proposed ScopeNet achieves state-of-the-art results on COCO
Semi-supervised learning (SSL) has a potential to improve the predictive performance of machine learning models using unlabeled data. Although there has been remarkable recent progress, the scope of demonstration in SSL has mainly been on image classification tasks. In this paper, we propose STAC, a simple yet effective SSL framework for visual object detection along with a data augmentation strategy. STAC deploys highly confident pseudo labels of localized objects from an unlabeled image and updates the model by enforcing consistency via strong augmentations. We propose experimental protocols to evaluate the performance of semi-supervised object detection using MS-COCO and show the efficacy of STAC on both MS-COCO and VOC07. On VOC07, STAC improves the AP$^{0.5}$ from $76.30$ to $79.08$; on MS-COCO, STAC demonstrates $2{\times}$ higher data efficiency by achieving 24.38 mAP using only 5\% labeled data than supervised baseline that marks 23.86\% using 10\% labeled data. The code is available at https://github.com/google-research/ssl_detection/.
Object-oriented maps are important for scene understanding since they jointly capture geometry and semantics, allow individual instantiation and meaningful reasoning about objects. We introduce FroDO, a method for accurate 3D reconstruction of object instances from RGB video that infers object location, pose and shape in a coarse-to-fine manner. Key to FroDO is to embed object shapes in a novel learnt space that allows seamless switching between sparse point cloud and dense DeepSDF decoding. Given an input sequence of localized RGB frames, FroDO first aggregates 2D detections to instantiate a category-aware 3D bounding box per object. A shape code is regressed using an encoder network before optimizing shape and pose further under the learnt shape priors using sparse and dense shape representations. The optimization uses multi-view geometric, photometric and silhouette losses. We evaluate on real-world datasets, including Pix3D, Redwood-OS, and ScanNet, for single-view, multi-view, and multi-object reconstruction.
In this paper, we present a method for simultaneous articulation model estimation and segmentation of an articulated object in RGB-D images using human hand motion. Our method uses the hand motion in the processes of the initial articulation model estimation, ICP-based model parameter optimization, and region selection of the target object. The hand motion gives an initial guess of the articulation model: prismatic or revolute joint. The method estimates the joint parameters by aligning the RGB-D images with the constraint of the hand motion. Finally, the target regions are selected from the cluster regions which move symmetrically along with the articulation model. Our experimental results show the robustness of the proposed method for the various objects.
Deep learning has revolutionized object detection thanks to large-scale datasets, but their object categories are still arguably very limited. In this paper, we attempt to enrich such categories by addressing the one-shot object detection problem, where the number of annotated training examples for learning an unseen class is limited to one. We introduce a two-stage model consisting of a first stage Matching-FCOS network and a second stage Structure-Aware Relation Module, the combination of which integrates metric learning with an anchor-free Faster R-CNN-style detection pipeline, eventually eliminating the need to fine-tune on the support images. We also propose novel training strategies that effectively improve detection performance. Extensive quantitative and qualitative evaluations were performed and our method exceeds the state-of-the-art one-shot performance consistently on multiple datasets.