PyTorchKR Papers

최신 논문

총 9,714편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2008.06941

Object-Aware Multi-Branch Relation Networks for Spatio-Temporal Video Grounding

Spatio-temporal video grounding aims to retrieve the spatio-temporal tube of a queried object according to the given sentence. Currently, most existing grounding methods are restricted to well-aligned segment-sentence pairs. In this paper, we explore spatio-temporal video grounding on unaligned data and multi-form sentences. This challenging task requires to capture critical object relations to identify the queried target. However, existing approaches cannot distinguish notable objects and remain in ineffective relation modeling between unnecessary objects. Thus, we propose a novel object-aware multi-branch relation network for object-aware relation discovery. Concretely, we first devise multiple branches to develop object-aware region modeling, where each branch focuses on a crucial object mentioned in the sentence. We then propose multi-branch relation reasoning to capture critical object relationships between the main branch and auxiliary branches. Moreover, we apply a diversity loss to make each branch only pay attention to its corresponding object and boost multi-branch learning. The extensive experiments show the effectiveness of our proposed method.

arXiv:2008.06828

A novel approach to remove foreign objects from chest X-ray images

We initially proposed a deep learning approach for foreign objects inpainting in smartphone-camera captured chest radiographs utilizing the cheXphoto dataset. Foreign objects which can significantly affect the quality of a computer-aided diagnostic prediction are captured under various settings. In this paper, we used multi-method to tackle both removal and inpainting chest radiographs. Firstly, an object detection model is trained to separate the foreign objects from the given image. Subsequently, the binary mask of each object is extracted utilizing a segmentation model. Each pair of the binary mask and the extracted object are then used for inpainting purposes. Finally, the in-painted regions are now merged back to the original image, resulting in a clean and non-foreign-object-existing output. To conclude, we achieved state-of-the-art accuracy. The experimental results showed a new approach to the possible applications of this method for chest X-ray images detection.

arXiv:2008.07294공식 코드 · Papers with Code

AP-Loss for Accurate One-Stage Object Detection

One-stage object detectors are trained by optimizing classification-loss and localization-loss simultaneously, with the former suffering much from extreme foreground-background class imbalance issue due to the large number of anchors. This paper alleviates this issue by proposing a novel framework to replace the classification task in one-stage detectors with a ranking task, and adopting the Average-Precision loss (AP-loss) for the ranking problem. Due to its non-differentiability and non-convexity, the AP-loss cannot be optimized directly. For this purpose, we develop a novel optimization algorithm, which seamlessly combines the error-driven update scheme in perceptron learning and backpropagation algorithm in deep networks. We provide in-depth analyses on the good convergence property and computational complexity of the proposed algorithm, both theoretically and empirically. Experimental results demonstrate notable improvement in addressing the imbalance issue in object detection over existing AP-based optimization algorithms. An improved state-of-the-art performance is achieved in one-stage detectors based on AP-loss over detectors using classification-losses on various standard benchmarks. The proposed framework is also highly versatile in accommodating different network architectures. Code is available at https://github.com/cccorn/AP-loss .

arXiv:2008.05955

SIDOD: A Synthetic Image Dataset for 3D Object Pose Recognition with Distractors

We present a new, publicly-available image dataset generated by the NVIDIA Deep Learning Data Synthesizer intended for use in object detection, pose estimation, and tracking applications. This dataset contains 144k stereo image pairs that synthetically combine 18 camera viewpoints of three photorealistic virtual environments with up to 10 objects (chosen randomly from the 21 object models of the YCB dataset [1]) and flying distractors. Object and camera pose, scene lighting, and quantity of objects and distractors were randomized. Each provided view includes RGB, depth, segmentation, and surface normal images, all pixel level. We describe our approach for domain randomization and provide insight into the decisions that produced the dataset.

arXiv:2008.05787

Shift Equivariance in Object Detection

Robustness to small image translations is a highly desirable property for object detectors. However, recent works have shown that CNN-based classifiers are not shift invariant. It is unclear to what extent this could impact object detection, mainly because of the architectural differences between the two and the dimensionality of the prediction space of modern detectors. To assess shift equivariance of object detection models end-to-end, in this paper we propose an evaluation metric, built upon a greedy search of the lower and upper bounds of the mean average precision on a shifted image set. Our new metric shows that modern object detection architectures, no matter if one-stage or two-stage, anchor-based or anchor-free, are sensitive to even one pixel shift to the input images. Furthermore, we investigate several possible solutions to this problem, both taken from the literature and newly proposed, quantifying the effectiveness of each one with the suggested metric. Our results indicate that none of these methods can provide full shift equivariance. Measuring and analyzing the extent of shift variance of different models and the contributions of possible factors, is a first step towards being able to devise methods that mitigate or even leverage such variabilities.

arXiv:2008.05700

What leads to generalization of object proposals?

Object proposal generation is often the first step in many detection models. It is lucrative to train a good proposal model, that generalizes to unseen classes. This could help scaling detection models to larger number of classes with fewer annotations. Motivated by this, we study how a detection model trained on a small set of source classes can provide proposals that generalize to unseen classes. We systematically study the properties of the dataset - visual diversity and label space granularity - required for good generalization. We show the trade-off between using fine-grained labels and coarse labels. We introduce the idea of prototypical classes: a set of sufficient and necessary classes required to train a detection model to obtain generalized proposals in a more data-efficient way. On the Open Images V4 dataset, we show that only 25% of the classes can be selected to form such a prototypical set. The resulting proposals from a model trained with these classes is only 4.3% worse than using all the classes, in terms of average recall (AR). We also demonstrate that Faster R-CNN model leads to better generalization of proposals compared to a single-stage network like RetinaNet.

arXiv:2008.05676공식 코드 · Papers with Code

Forest R-CNN: Large-Vocabulary Long-Tailed Object Detection and Instance Segmentation

Despite the previous success of object analysis, detecting and segmenting a large number of object categories with a long-tailed data distribution remains a challenging problem and is less investigated. For a large-vocabulary classifier, the chance of obtaining noisy logits is much higher, which can easily lead to a wrong recognition. In this paper, we exploit prior knowledge of the relations among object categories to cluster fine-grained classes into coarser parent classes, and construct a classification tree that is responsible for parsing an object instance into a fine-grained category via its parent class. In the classification tree, as the number of parent class nodes are significantly less, their logits are less noisy and can be utilized to suppress the wrong/noisy logits existed in the fine-grained class nodes. As the way to construct the parent class is not unique, we further build multiple trees to form a classification forest where each tree contributes its vote to the fine-grained classification. To alleviate the imbalanced learning caused by the long-tail phenomena, we propose a simple yet effective resampling method, NMS Resampling, to re-balance the data distribution. Our method, termed as Forest R-CNN, can serve as a plug-and-play module being applied to most object recognition models for recognizing more than 1000 categories. Extensive experiments are performed on the large vocabulary dataset LVIS. Compared with the Mask R-CNN baseline, the Forest R-CNN significantly boosts the performance with 11.5% and 3.9% AP improvements on the rare categories and overall categories, respectively. Moreover, we achieve state-of-the-art results on the LVIS dataset. Code is available at https://github.com/JialianW/Forest_RCNN.

arXiv:2008.05534

Co-training for On-board Deep Object Detection

Providing ground truth supervision to train visual models has been a bottleneck over the years, exacerbated by domain shifts which degenerate the performance of such models. This was the case when visual tasks relied on handcrafted features and shallow machine learning and, despite its unprecedented performance gains, the problem remains open within the deep learning paradigm due to its data-hungry nature. Best performing deep vision-based object detectors are trained in a supervised manner by relying on human-labeled bounding boxes which localize class instances (i.e.objects) within the training images.Thus, object detection is one of such tasks for which human labeling is a major bottleneck. In this paper, we assess co-training as a semi-supervised learning method for self-labeling objects in unlabeled images, so reducing the human-labeling effort for developing deep object detectors. Our study pays special attention to a scenario involving domain shift; in particular, when we have automatically generated virtual-world images with object bounding boxes and we have real-world images which are unlabeled. Moreover, we are particularly interested in using co-training for deep object detection in the context of driver assistance systems and/or self-driving vehicles. Thus, using well-established datasets and protocols for object detection in these application contexts, we will show how co-training is a paradigm worth to pursue for alleviating object labeling, working both alone and together with task-agnostic domain adaptation.

arXiv:2008.05397

Rethinking of the Image Salient Object Detection: Object-level Semantic Saliency Re-ranking First, Pixel-wise Saliency Refinement Latter

The real human attention is an interactive activity between our visual system and our brain, using both low-level visual stimulus and high-level semantic information. Previous image salient object detection (SOD) works conduct their saliency predictions in a multi-task manner, i.e., performing pixel-wise saliency regression and segmentation-like saliency refinement at the same time, which degenerates their feature backbones in revealing semantic information. However, given an image, we tend to pay more attention to those regions which are semantically salient even in the case that these regions are perceptually not the most salient ones at first glance. In this paper, we divide the SOD problem into two sequential tasks: 1) we propose a lightweight, weakly supervised deep network to coarsely locate those semantically salient regions first; 2) then, as a post-processing procedure, we selectively fuse multiple off-the-shelf deep models on these semantically salient regions as the pixel-wise saliency refinement. In sharp contrast to the state-of-the-art (SOTA) methods that focus on learning pixel-wise saliency in "single image" using perceptual clues mainly, our method has investigated the "object-level semantic ranks between multiple images", of which the methodology is more consistent with the real human attention mechanism. Our method is simple yet effective, which is the first attempt to consider the salient object detection mainly as an object-level semantic re-ranking problem.

arXiv:2008.05096공식 코드 · Papers with Code

Inter-Image Communication for Weakly Supervised Localization

Weakly supervised localization aims at finding target object regions using only image-level supervision. However, localization maps extracted from classification networks are often not accurate due to the lack of fine pixel-level supervision. In this paper, we propose to leverage pixel-level similarities across different objects for learning more accurate object locations in a complementary way. Particularly, two kinds of constraints are proposed to prompt the consistency of object features within the same categories. The first constraint is to learn the stochastic feature consistency among discriminative pixels that are randomly sampled from different images within a batch. The discriminative information embedded in one image can be leveraged to benefit its counterpart with inter-image communication. The second constraint is to learn the global consistency of object features throughout the entire dataset. We learn a feature center for each category and realize the global feature consistency by forcing the object features to approach class-specific centers. The global centers are actively updated with the training process. The two constraints can benefit each other to learn consistent pixel-level features within the same categories, and finally improve the quality of localization maps. We conduct extensive experiments on two popular benchmarks, i.e., ILSVRC and CUB-200-2011. Our method achieves the Top-1 localization error rate of 45.17% on the ILSVRC validation set, surpassing the current state-of-the-art method by a large margin. The code is available at https://github.com/xiaomengyc/I2C.

arXiv:2008.05242

PAM:Point-wise Attention Module for 6D Object Pose Estimation

6D pose estimation refers to object recognition and estimation of 3D rotation and 3D translation. The key technology for estimating 6D pose is to estimate pose by extracting enough features to find pose in any environment. Previous methods utilized depth information in the refinement process or were designed as a heterogeneous architecture for each data space to extract feature. However, these methods are limited in that they cannot extract sufficient feature. Therefore, this paper proposes a Point Attention Module that can efficiently extract powerful feature from RGB-D. In our Module, attention map is formed through a Geometric Attention Path(GAP) and Channel Attention Path(CAP). In GAP, it is designed to pay attention to important information in geometric information, and CAP is designed to pay attention to important information in Channel information. We show that the attention module efficiently creates feature representations without significantly increasing computational complexity. Experimental results show that the proposed method outperforms the existing methods in benchmarks, YCB Video and LineMod. In addition, the attention module was applied to the classification task, and it was confirmed that the performance significantly improved compared to the existing model.

arXiv:2008.04465

Safe and Effective Picking Paths in Clutter given Discrete Distributions of Object Poses

Picking an item in the presence of other objects can be challenging as it involves occlusions and partial views. Given object models, one approach is to perform object pose estimation and use the most likely candidate pose per object to pick the target without collisions. This approach, however, ignores the uncertainty of the perception process both regarding the target's and the surrounding objects' poses. This work proposes first a perception process for 6D pose estimation, which returns a discrete distribution of object poses in a scene. Then, an open-loop planning pipeline is proposed to return safe and effective solutions for moving a robotic arm to pick, which (a) minimizes the probability of collision with the obstructing objects; and (b) maximizes the probability of reaching the target item. The planning framework models the challenge as a stochastic variant of the Minimum Constraint Removal (MCR) problem. The effectiveness of the methodology is verified given both simulated and real data in different scenarios. The experiments demonstrate the importance of considering the uncertainty of the perception process in terms of safe execution. The results also show that the methodology is more effective than conservative MCR approaches, which avoid all possible object poses regardless of the reported uncertainty.

arXiv:2008.04451공식 코드 · Papers with Code

Grasping Field: Learning Implicit Representations for Human Grasps

Robotic grasping of house-hold objects has made remarkable progress in recent years. Yet, human grasps are still difficult to synthesize realistically. There are several key reasons: (1) the human hand has many degrees of freedom (more than robotic manipulators); (2) the synthesized hand should conform to the surface of the object; and (3) it should interact with the object in a semantically and physically plausible manner. To make progress in this direction, we draw inspiration from the recent progress on learning-based implicit representations for 3D object reconstruction. Specifically, we propose an expressive representation for human grasp modelling that is efficient and easy to integrate with deep neural networks. Our insight is that every point in a three-dimensional space can be characterized by the signed distances to the surface of the hand and the object, respectively. Consequently, the hand, the object, and the contact area can be represented by implicit surfaces in a common space, in which the proximity between the hand and the object can be modelled explicitly. We name this 3D to 2D mapping as Grasping Field, parameterize it with a deep neural network, and learn it from data. We demonstrate that the proposed grasping field is an effective and expressive representation for human grasp generation. Specifically, our generative model is able to synthesize high-quality human grasps, given only on a 3D object point cloud. The extensive experiments demonstrate that our generative model compares favorably with a strong baseline and approaches the level of natural human grasps. Our method improves the physical plausibility of the hand-object contact reconstruction and achieves comparable performance for 3D hand reconstruction compared to state-of-the-art methods.

arXiv:2008.04738

Attention-based 3D Object Reconstruction from a Single Image

Recently, learning-based approaches for 3D reconstruction from 2D images have gained popularity due to its modern applications, e.g., 3D printers, autonomous robots, self-driving cars, virtual reality, and augmented reality. The computer vision community has applied a great effort in developing functions to reconstruct the full 3D geometry of objects and scenes. However, to extract image features, they rely on convolutional neural networks, which are ineffective in capturing long-range dependencies. In this paper, we propose to substantially improve Occupancy Networks, a state-of-the-art method for 3D object reconstruction. For such we apply the concept of self-attention within the network's encoder in order to leverage complementary input features rather than those based on local regions, helping the encoder to extract global information. With our approach, we were capable of improving the original work in 5.05% of mesh IoU, 0.83% of Normal Consistency, and more than 10X the Chamfer-L1 distance. We also perform a qualitative study that shows that our approach was able to generate much more consistent meshes, confirming its increased generalization power over the current state-of-the-art.

arXiv:2008.04159

Depth Quality Aware Salient Object Detection

The existing fusion based RGB-D salient object detection methods usually adopt the bi-stream structure to strike the fusion trade-off between RGB and depth (D). The D quality usually varies from scene to scene, while the SOTA bi-stream approaches are depth quality unaware, which easily result in substantial difficulties in achieving complementary fusion status between RGB and D, leading to poor fusion results in facing of low-quality D. Thus, this paper attempts to integrate a novel depth quality aware subnet into the classic bi-stream structure, aiming to assess the depth quality before conducting the selective RGB-D fusion. Compared with the SOTA bi-stream methods, the major highlight of our method is its ability to lessen the importance of those low-quality, no-contribution, or even negative-contribution D regions during the RGB-D fusion, achieving a much improved complementary status between RGB and D.

arXiv:2008.04094

Adversarial Examples on Object Recognition: A Comprehensive Survey

Deep neural networks are at the forefront of machine learning research. However, despite achieving impressive performance on complex tasks, they can be very sensitive: Small perturbations of inputs can be sufficient to induce incorrect behavior. Such perturbations, called adversarial examples, are intentionally designed to test the network's sensitivity to distribution drifts. Given their surprisingly small size, a wide body of literature conjectures on their existence and how this phenomenon can be mitigated. In this article we discuss the impact of adversarial examples on security, safety, and robustness of neural networks. We start by introducing the hypotheses behind their existence, the methods used to construct or protect against them, and the capacity to transfer adversarial examples between different machine learning models. Altogether, the goal is to provide a comprehensive and self-contained survey of this growing field of research.

arXiv:2008.03511

Single-Shot Two-Pronged Detector with Rectified IoU Loss

In the CNN based object detectors, feature pyramids are widely exploited to alleviate the problem of scale variation across object instances. These object detectors, which strengthen features via a top-down pathway and lateral connections, are mainly to enrich the semantic information of low-level features, but ignore the enhancement of high-level features. This can lead to an imbalance between different levels of features, in particular a serious lack of detailed information in the high-level features, which makes it difficult to get accurate bounding boxes. In this paper, we introduce a novel two-pronged transductive idea to explore the relationship among different layers in both backward and forward directions, which can enrich the semantic information of low-level features and detailed information of high-level features at the same time. Under the guidance of the two-pronged idea, we propose a Two-Pronged Network (TPNet) to achieve bidirectional transfer between high-level features and low-level features, which is useful for accurately detecting object at different scales. Furthermore, due to the distribution imbalance between the hard and easy samples in single-stage detectors, the gradient of localization loss is always dominated by the hard examples that have poor localization accuracy. This will enable the model to be biased toward the hard samples. So in our TPNet, an adaptive IoU based localization loss, named Rectified IoU (RIoU) loss, is proposed to rectify the gradients of each kind of samples. The Rectified IoU loss increases the gradients of examples with high IoU while suppressing the gradients of examples with low IoU, which can improve the overall localization accuracy of model. Extensive experiments demonstrate the superiority of our TPNet and RIoU loss.

arXiv:2008.03346

Generative Adversarial Network for Radar Signal Generation

A major obstacle in radar based methods for concealed object detection on humans and seamless integration into security and access control system is the difficulty in collecting high quality radar signal data. Generative adversarial networks (GAN) have shown promise in data generation application in the fields of image and audio processing. As such, this paper proposes the design of a GAN for application in radar signal generation. Data collected using the Finite-Difference Time-Domain (FDTD) method on three concealed object classes (no object, large object, and small object) were used as training data to train a GAN to generate radar signal samples for each class. The proposed GAN generated radar signal data which was indistinguishable from the training data by qualitative human observers.

arXiv:2008.03718

1-Point RANSAC-Based Method for Ground Object Pose Estimation

Solving Perspective-n-Point (PnP) problems is a traditional way of estimating object poses. Given outlier-contaminated data, a pose of an object is calculated with PnP algorithms of n = {3, 4} in the RANSAC-based scheme. However, the computational complexity considerably increases along with n and the high complexity imposes a severe strain on devices which should estimate multiple object poses in real time. In this paper, we propose an efficient method based on 1-point RANSAC for estimating a pose of an object on the ground. In the proposed method, a pose is calculated with 1-DoF parameterization by using a ground object assumption and a 2D object bounding box as an additional observation, thereby achieving the fastest performance among the RANSAC-based methods. In addition, since the method suffers from the errors of the additional information, we propose a hierarchical robust estimation method for polishing a rough pose estimate and discovering more inliers in a coarse-to-fine manner. The experiments in synthetic and real-world datasets demonstrate the superiority of the proposed method.

arXiv:2008.03696

Radar-based Dynamic Occupancy Grid Mapping and Object Detection

Environment modeling utilizing sensor data fusion and object tracking is crucial for safe automated driving. In recent years, the classical occupancy grid map approach, which assumes a static environment, has been extended to dynamic occupancy grid maps, which maintain the possibility of a low-level data fusion while also estimating the position and velocity distribution of the dynamic local environment. This paper presents the further development of a previous approach. To the best of the author's knowledge, there is no publication about dynamic occupancy grid mapping with subsequent analysis based only on radar data. Therefore in this work, the data of multiple radar sensors are fused, and a grid-based object tracking and mapping method is applied. Subsequently, the clustering of dynamic areas provides high-level object information. For comparison, also a lidar-based method is developed. The approach is evaluated qualitatively and quantitatively with real-world data from a moving vehicle in urban environments. The evaluation illustrates the advantages of the radar-based dynamic occupancy grid map, considering different comparison metrics.