PyTorchKR Papers

최신 논문

총 9,714편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2012.00313

Unsupervised Part Discovery via Feature Alignment

Understanding objects in terms of their individual parts is important, because it enables a precise understanding of the objects' geometrical structure, and enhances object recognition when the object is seen in a novel pose or under partial occlusion. However, the manual annotation of parts in large scale datasets is time consuming and expensive. In this paper, we aim at discovering object parts in an unsupervised manner, i.e., without ground-truth part or keypoint annotations. Our approach builds on the intuition that objects of the same class in a similar pose should have their parts aligned at similar spatial locations. We exploit the property that neural network features are largely invariant to nuisance variables and the main remaining source of variations between images of the same object category is the object pose. Specifically, given a training image, we find a set of similar images that show instances of the same object category in the same pose, through an affine alignment of their corresponding feature maps. The average of the aligned feature maps serves as pseudo ground-truth annotation for a supervised training of the deep network backbone. During inference, part detection is simple and fast, without any extra modules or overheads other than a feed-forward neural network. Our experiments on several datasets from different domains verify the effectiveness of the proposed method. For example, we achieve 37.8 mAP on VehiclePart, which is at least 4.2 better than previous methods.

arXiv:2012.00057공식 코드 · Papers with Code

Move to See Better: Self-Improving Embodied Object Detection

Passive methods for object detection and segmentation treat images of the same scene as individual samples and do not exploit object permanence across multiple views. Generalization to novel or difficult viewpoints thus requires additional training with lots of annotations. In contrast, humans often recognize objects by simply moving around, to get more informative viewpoints. In this paper, we propose a method for improving object detection in testing environments, assuming nothing but an embodied agent with a pre-trained 2D object detector. Our agent collects multi-view data, generates 2D and 3D pseudo-labels, and fine-tunes its detector in a self-supervised manner. Experiments on both indoor and outdoor datasets show that (1) our method obtains high-quality 2D and 3D pseudo-labels from multi-view RGB-D data; (2) fine-tuning with these pseudo-labels improves the 2D detector significantly in the test environment; (3) training a 3D detector with our pseudo-labels outperforms a prior self-supervised method by a large margin; (4) given weak supervision, our method can generate better pseudo-labels for novel objects.

arXiv:2011.13977

Improving Welfare in One-sided Matching using Simple Threshold Queries

We study one-sided matching problems where $n$ agents have preferences over $m$ objects and each of them need to be assigned to at most one object. Most work on such problems assume that the agents only have ordinal preferences and usually the goal in them is to compute a matching that satisfies some notion of economic efficiency. However, in reality, agents may have some preference intensities or cardinal utilities that, e.g., indicate that they like an object much more than another object, and not taking these into account can result in a loss in welfare. While one way to potentially account for these is to directly ask the agents for this information, such an elicitation process is cognitively demanding. Therefore, we focus on learning more about their cardinal preferences using simple threshold queries which ask an agent if they value an object greater than a certain value, and use this in turn to come up with algorithms that produce a matching that, for a particular economic notion $X$, satisfies $X$ and also achieves a good approximation to the optimal welfare among all matchings that satisfy $X$. We focus on several notions of economic efficiency, and look at both adaptive and non-adaptive algorithms. Overall, our results show how one can improve welfare by even non-adaptively asking the agents for just one bit of extra information per object.

arXiv:2011.14381공식 코드 · Papers with Code

Self-supervised Visual Reinforcement Learning with Object-centric Representations

Autonomous agents need large repertoires of skills to act reasonably on new tasks that they have not seen before. However, acquiring these skills using only a stream of high-dimensional, unstructured, and unlabeled observations is a tricky challenge for any autonomous agent. Previous methods have used variational autoencoders to encode a scene into a low-dimensional vector that can be used as a goal for an agent to discover new skills. Nevertheless, in compositional/multi-object environments it is difficult to disentangle all the factors of variation into such a fixed-length representation of the whole scene. We propose to use object-centric representations as a modular and structured observation space, which is learned with a compositional generative world model. We show that the structure in the representations in combination with goal-conditioned attention policies helps the autonomous agent to discover and learn useful skills. These skills can be further combined to address compositional tasks like the manipulation of several different objects.

arXiv:2011.14289

Learning geometry-image representation for 3D point cloud generation

We study the problem of generating point clouds of 3D objects. Instead of discretizing the object into 3D voxels with huge computational cost and resolution limitations, we propose a novel geometry image based generator (GIG) to convert the 3D point cloud generation problem to a 2D geometry image generation problem. Since the geometry image is a completely regular 2D array that contains the surface points of the 3D object, it leverages both the regularity of the 2D array and the geodesic neighborhood of the 3D surface. Thus, one significant benefit of our GIG is that it allows us to directly generate the 3D point clouds using efficient 2D image generation networks. Experiments on both rigid and non-rigid 3D object datasets have demonstrated the promising performance of our method to not only create plausible and novel 3D objects, but also learn a probabilistic latent space that well supports the shape editing like interpolation and arithmetic.

arXiv:2011.14204

Class-agnostic Object Detection

Object detection models perform well at localizing and classifying objects that they are shown during training. However, due to the difficulty and cost associated with creating and annotating detection datasets, trained models detect a limited number of object types with unknown objects treated as background content. This hinders the adoption of conventional detectors in real-world applications like large-scale object matching, visual grounding, visual relation prediction, obstacle detection (where it is more important to determine the presence and location of objects than to find specific types), etc. We propose class-agnostic object detection as a new problem that focuses on detecting objects irrespective of their object-classes. Specifically, the goal is to predict bounding boxes for all objects in an image but not their object-classes. The predicted boxes can then be consumed by another system to perform application-specific classification, retrieval, etc. We propose training and evaluation protocols for benchmarking class-agnostic detectors to advance future research in this domain. Finally, we propose (1) baseline methods and (2) a new adversarial learning framework for class-agnostic detection that forces the model to exclude class-specific information from features used for predictions. Experimental results show that adversarial learning improves class-agnostic detection efficacy.

arXiv:2011.14744공식 코드 · Papers with Code

RfD-Net: Point Scene Understanding by Semantic Instance Reconstruction

Semantic scene understanding from point clouds is particularly challenging as the points reflect only a sparse set of the underlying 3D geometry. Previous works often convert point cloud into regular grids (e.g. voxels or bird-eye view images), and resort to grid-based convolutions for scene understanding. In this work, we introduce RfD-Net that jointly detects and reconstructs dense object surfaces directly from raw point clouds. Instead of representing scenes with regular grids, our method leverages the sparsity of point cloud data and focuses on predicting shapes that are recognized with high objectness. With this design, we decouple the instance reconstruction into global object localization and local shape prediction. It not only eases the difficulty of learning 2-D manifold surfaces from sparse 3D space, the point clouds in each object proposal convey shape details that support implicit function learning to reconstruct any high-resolution surfaces. Our experiments indicate that instance detection and reconstruction present complementary effects, where the shape prediction head shows consistent effects on improving object detection with modern 3D proposal network backbones. The qualitative and quantitative evaluations further demonstrate that our approach consistently outperforms the state-of-the-arts and improves over 11 of mesh IoU in object reconstruction.

arXiv:2011.14589공식 코드 · Papers with Code

Monocular 3D Object Detection with Sequential Feature Association and Depth Hint Augmentation

Monocular 3D object detection, with the aim of predicting the geometric properties of on-road objects, is a promising research topic for the intelligent perception systems of autonomous driving. Most state-of-the-art methods follow a keypoint-based paradigm, where the keypoints of objects are predicted and employed as the basis for regressing the other geometric properties. In this work, a unified network named as FADNet is presented to address the task of monocular 3D object detection. In contrast to previous keypoint-based methods, we propose to divide the output modalities into different groups according to the estimation difficulty of object properties. Different groups are treated differently and sequentially associated by a convolutional Gated Recurrent Unit. Another contribution of this work is the strategy of depth hint augmentation. To provide characterized depth patterns as hints for depth estimation, a dedicated depth hint module is designed to generate row-wise features named as depth hints, which are explicitly supervised in a bin-wise manner. The contributions of this work are validated by conducting experiments and ablation study on the KITTI benchmark. Without utilizing depth priors, post optimization, or other refinement modules, our network performs competitively against state-of-the-art methods while maintaining a decent running speed.

arXiv:2011.14448

Improved Handling of Motion Blur in Online Object Detection

We wish to detect specific categories of objects, for online vision systems that will run in the real world. Object detection is already very challenging. It is even harder when the images are blurred, from the camera being in a car or a hand-held phone. Most existing efforts either focused on sharp images, with easy to label ground truth, or they have treated motion blur as one of many generic corruptions. Instead, we focus especially on the details of egomotion induced blur. We explore five classes of remedies, where each targets different potential causes for the performance gap between sharp and blurred images. For example, first deblurring an image changes its human interpretability, but at present, only partly improves object detection. The other four classes of remedies address multi-scale texture, out-of-distribution testing, label generation, and conditioning by blur-type. Surprisingly, we discover that custom label generation aimed at resolving spatial ambiguity, ahead of all others, markedly improves object detection. Also, in contrast to findings from classification, we see a noteworthy boost by conditioning our model on bespoke categories of motion blur. We validate and cross-breed the different remedies experimentally on blurred COCO images and real-world blur datasets, producing an easy and practical favorite model with superior detection rates.

arXiv:2011.15067

Learning a metacognition for object perception

Beyond representing the external world, humans also represent their own cognitive processes. In the context of perception, this metacognition helps us identify unreliable percepts, such as when we recognize that we are seeing an illusion. Here we propose MetaGen, a model for the unsupervised learning of metacognition. In MetaGen, metacognition is expressed as a generative model of how a perceptual system produces noisy percepts. Using basic principles of how the world works (such as object permanence, part of infants' core knowledge), MetaGen jointly infers the objects in the world causing the percepts and a representation of its own perceptual system. MetaGen can then use this metacognition to infer which objects are actually present in the world. On simulated data, we find that MetaGen quickly learns a metacognition and improves overall accuracy, outperforming models that lack a metacognition.

arXiv:2011.13075

Energy Drain of the Object Detection Processing Pipeline for Mobile Devices: Analysis and Implications

Applying deep learning to object detection provides the capability to accurately detect and classify complex objects in the real world. However, currently, few mobile applications use deep learning because such technology is computation-intensive and energy-consuming. This paper, to the best of our knowledge, presents the first detailed experimental study of a mobile augmented reality (AR) client's energy consumption and the detection latency of executing Convolutional Neural Networks (CNN) based object detection, either locally on the smartphone or remotely on an edge server. In order to accurately measure the energy consumption on the smartphone and obtain the breakdown of energy consumed by each phase of the object detection processing pipeline, we propose a new measurement strategy. Our detailed measurements refine the energy analysis of mobile AR clients and reveal several interesting perspectives regarding the energy consumption of executing CNN-based object detection. Furthermore, several insights and research opportunities are proposed based on our experimental results. These findings from our experimental study will guide the design of energy-efficient processing pipeline of CNN-based object detection.

arXiv:2011.13375공식 코드 · Papers with Code

Invisible Perturbations: Physical Adversarial Examples Exploiting the Rolling Shutter Effect

Physical adversarial examples for camera-based computer vision have so far been achieved through visible artifacts -- a sticker on a Stop sign, colorful borders around eyeglasses or a 3D printed object with a colorful texture. An implicit assumption here is that the perturbations must be visible so that a camera can sense them. By contrast, we contribute a procedure to generate, for the first time, physical adversarial examples that are invisible to human eyes. Rather than modifying the victim object with visible artifacts, we modify light that illuminates the object. We demonstrate how an attacker can craft a modulated light signal that adversarially illuminates a scene and causes targeted misclassifications on a state-of-the-art ImageNet deep learning model. Concretely, we exploit the radiometric rolling shutter effect in commodity cameras to create precise striping patterns that appear on images. To human eyes, it appears like the object is illuminated, but the camera creates an image with stripes that will cause ML models to output the attacker-desired classification. We conduct a range of simulation and physical experiments with LEDs, demonstrating targeted attack rates up to 84%.

arXiv:2011.13209

Handling Object Symmetries in CNN-based Pose Estimation

In this paper, we investigate the problems that Convolutional Neural Networks (CNN)-based pose estimators have with symmetric objects. We considered the value of the CNN's output representation when continuously rotating the object and found that it has to form a closed loop after each step of symmetry. Otherwise, the CNN (which is itself a continuous function) has to replicate an uncontinuous function. On a 1-DOF toy example we show that commonly used representations do not fulfill this demand and analyze the problems caused thereby. In particular, we find that the popular min-over-symmetries approach for creating a symmetry-aware loss tends not to work well with gradient-based optimization, i.e. deep learning. We propose a representation called "closed symmetry loop" (csl) from these insights, where the angle of relevant vectors is multiplied by the symmetry order and then generalize it to 6-DOF. The representation extends our algorithm from [Richter-Klug, ICVS, 2019] including a method to disambiguate symmetric equivalents during the final pose estimation. The algorithm handles continuous rotational symmetry (e.g. a bottle) and discrete rotational symmetry (e.g. a 4-fold symmetric box). It is evaluated on the T-LESS dataset, where it reaches state-of-the-art for unrefining RGB-based methods.

arXiv:2011.13843공식 코드 · Papers with Code

SFTrack++: A Fast Learnable Spectral Segmentation Approach for Space-Time Consistent Tracking

We propose an object tracking method, SFTrack++, that smoothly learns to preserve the tracked object consistency over space and time dimensions by taking a spectral clustering approach over the graph of pixels from the video, using a fast 3D filtering formulation for finding the principal eigenvector of this graph's adjacency matrix. To better capture complex aspects of the tracked object, we enrich our formulation to multi-channel inputs, which permit different points of view for the same input. The channel inputs are in our experiments, the output of multiple tracking methods. After combining them, instead of relying only on hidden layers representations to predict a good tracking bounding box, we explicitly learn an intermediate, more refined one, namely the segmentation map of the tracked object. This prevents the rough common bounding box approach to introduce noise and distractors in the learning process. We test our method, SFTrack++, on five tracking benchmarks: OTB, UAV, NFS, GOT-10k, and TrackingNet, using five top trackers as input. Our experimental results validate the pre-registered hypothesis. We obtain consistent and robust results, competitive on the three traditional benchmarks (OTB, UAV, NFS) and significantly on top of others (by over $1.1\%$ on accuracy) on GOT-10k and TrackingNet, which are newer, larger, and more varied datasets.

arXiv:2011.13677

Self-EMD: Self-Supervised Object Detection without ImageNet

In this paper, we propose a novel self-supervised representation learning method, Self-EMD, for object detection. Our method directly trained on unlabeled non-iconic image dataset like COCO, instead of commonly used iconic-object image dataset like ImageNet. We keep the convolutional feature maps as the image embedding to preserve spatial structures and adopt Earth Mover's Distance (EMD) to compute the similarity between two embeddings. Our Faster R-CNN (ResNet50-FPN) baseline achieves 39.8% mAP on COCO, which is on par with the state of the art self-supervised methods pre-trained on ImageNet. More importantly, it can be further improved to 40.4% mAP with more unlabeled images, showing its great potential for leveraging more easily obtained unlabeled data. Code will be made available.

arXiv:2011.13669

Towards real-time object recognition and pose estimation in point clouds

Object recognition and 6DoF pose estimation are quite challenging tasks in computer vision applications. Despite efficiency in such tasks, standard methods deliver far from real-time processing rates. This paper presents a novel pipeline to estimate a fine 6DoF pose of objects, applied to realistic scenarios in real-time. We split our proposal into three main parts. Firstly, a Color feature classification leverages the use of pre-trained CNN color features trained on the ImageNet for object detection. A Feature-based registration module conducts a coarse pose estimation, and finally, a Fine-adjustment step performs an ICP-based dense registration. Our proposal achieves, in the best case, an accuracy performance of almost 83\% on the RGB-D Scenes dataset. Regarding processing time, the object detection task is done at a frame processing rate up to 90 FPS, and the pose estimation at almost 14 FPS in a full execution strategy. We discuss that due to the proposal's modularity, we could let the full execution occurs only when necessary and perform a scheduled execution that unlocks real-time processing, even for multitask situations.

arXiv:2011.13628

Temporal-Channel Transformer for 3D Lidar-Based Video Object Detection in Autonomous Driving

The strong demand of autonomous driving in the industry has lead to strong interest in 3D object detection and resulted in many excellent 3D object detection algorithms. However, the vast majority of algorithms only model single-frame data, ignoring the temporal information of the sequence of data. In this work, we propose a new transformer, called Temporal-Channel Transformer, to model the spatial-temporal domain and channel domain relationships for video object detecting from Lidar data. As a special design of this transformer, the information encoded in the encoder is different from that in the decoder, i.e. the encoder encodes temporal-channel information of multiple frames while the decoder decodes the spatial-channel information for the current frame in a voxel-wise manner. Specifically, the temporal-channel encoder of the transformer is designed to encode the information of different channels and frames by utilizing the correlation among features from different channels and frames. On the other hand, the spatial decoder of the transformer will decode the information for each location of the current frame. Before conducting the object detection with detection head, the gate mechanism is deployed for re-calibrating the features of current frame, which filters out the object irrelevant information by repetitively refine the representation of target frame along with the up-sampling process. Experimental results show that we achieve the state-of-the-art performance in grid voxel-based 3D object detection on the nuScenes benchmark.

arXiv:2011.13560

SocialGuard: An Adversarial Example Based Privacy-Preserving Technique for Social Images

The popularity of various social platforms has prompted more people to share their routine photos online. However, undesirable privacy leakages occur due to such online photo sharing behaviors. Advanced deep neural network (DNN) based object detectors can easily steal users' personal information exposed in shared photos. In this paper, we propose a novel adversarial example based privacy-preserving technique for social images against object detectors based privacy stealing. Specifically, we develop an Object Disappearance Algorithm to craft two kinds of adversarial social images. One can hide all objects in the social images from being detected by an object detector, and the other can make the customized sensitive objects be incorrectly classified by the object detector. The Object Disappearance Algorithm constructs perturbation on a clean social image. After being injected with the perturbation, the social image can easily fool the object detector, while its visual quality will not be degraded. We use two metrics, privacy-preserving success rate and privacy leakage rate, to evaluate the effectiveness of the proposed method. Experimental results show that, the proposed method can effectively protect the privacy of social images. The privacy-preserving success rates of the proposed method on MS-COCO and PASCAL VOC 2007 datasets are high up to 96.1% and 99.3%, respectively, and the privacy leakage rates on these two datasets are as low as 0.57% and 0.07%, respectively. In addition, compared with existing image processing methods (low brightness, noise, blur, mosaic and JPEG compression), the proposed method can achieve much better performance in privacy protection and image visual quality maintenance.

arXiv:2011.12550

Robust Correlation Tracking via Multi-channel Fused Features and Reliable Response Map

Benefiting from its ability to efficiently learn how an object is changing, correlation filters have recently demonstrated excellent performance for rapidly tracking objects. Designing effective features and handling model drifts are two important aspects for online visual tracking. This paper tackles these challenges by proposing a robust correlation tracking algorithm (RCT) based on two ideas: First, we propose a method to fuse features in order to more naturally describe the gradient and color information of the tracked object, and introduce the fused features into a background aware correlation filter to obtain the response map. Second, we present a novel strategy to significantly reduce noise in the response map and therefore ease the problem of model drift. Systematic comparative evaluations performed over multiple tracking benchmarks demonstrate the efficacy of the proposed approach.

arXiv:2011.12450공식 코드 · Papers with Code

Sparse R-CNN: End-to-End Object Detection with Learnable Proposals

We present Sparse R-CNN, a purely sparse method for object detection in images. Existing works on object detection heavily rely on dense object candidates, such as $k$ anchor boxes pre-defined on all grids of image feature map of size $H\times W$. In our method, however, a fixed sparse set of learned object proposals, total length of $N$, are provided to object recognition head to perform classification and location. By eliminating $HWk$ (up to hundreds of thousands) hand-designed object candidates to $N$ (e.g. 100) learnable proposals, Sparse R-CNN completely avoids all efforts related to object candidates design and many-to-one label assignment. More importantly, final predictions are directly output without non-maximum suppression post-procedure. Sparse R-CNN demonstrates accuracy, run-time and training convergence performance on par with the well-established detector baselines on the challenging COCO dataset, e.g., achieving 45.0 AP in standard $3\times$ training schedule and running at 22 fps using ResNet-50 FPN model. We hope our work could inspire re-thinking the convention of dense prior in object detectors. The code is available at: https://github.com/PeizeSun/SparseR-CNN.