PyTorchKR Papers

최신 논문

총 9,714편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2003.09150공식 코드 · Papers with Code

Multi-branch and Multi-scale Attention Learning for Fine-Grained Visual Categorization

ImageNet Large Scale Visual Recognition Challenge (ILSVRC) is one of the most authoritative academic competitions in the field of Computer Vision (CV) in recent years. But applying ILSVRC's annual champion directly to fine-grained visual categorization (FGVC) tasks does not achieve good performance. To FGVC tasks, the small inter-class variations and the large intra-class variations make it a challenging problem. Our attention object location module (AOLM) can predict the position of the object and attention part proposal module (APPM) can propose informative part regions without the need of bounding-box or part annotations. The obtained object images not only contain almost the entire structure of the object, but also contains more details, part images have many different scales and more fine-grained features, and the raw images contain the complete object. The three kinds of training images are supervised by our multi-branch network. Therefore, our multi-branch and multi-scale learning network(MMAL-Net) has good classification ability and robustness for images of different scales. Our approach can be trained end-to-end, while provides short inference time. Through the comprehensive experiments demonstrate that our approach can achieves state-of-the-art results on CUB-200-2011, FGVC-Aircraft and Stanford Cars datasets. Our code will be available at https://github.com/ZF1044404254/MMAL-Net

arXiv:2003.08771

"Who is Driving around Me?" Unique Vehicle Instance Classification using Deep Neural Features

Being aware of other traffic is a prerequisite for self-driving cars to operate in the real world. In this paper, we show how the intrinsic feature maps of an object detection CNN can be used to uniquely identify vehicles from a dash-cam feed. Feature maps of a pretrained `YOLO' network are used to create 700 deep integrated feature signatures (DIFS) from 20 different images of 35 vehicles from a high resolution dataset and 340 signatures from 20 different images of 17 vehicles of a lower resolution tracking benchmark dataset. The YOLO network was trained to classify general object categories, e.g. classify a detected object as a `car' or `truck'. 5-Fold nearest neighbor (1NN) classification was used on DIFS created from feature maps in the middle layers of the network to correctly identify unique vehicles at a rate of 96.7\% for the high resolution data and with a rate of 86.8\% for the lower resolution data. We conclude that a deep neural detection network trained to distinguish between different classes can be successfully used to identify different instances belonging to the same class, through the creation of deep integrated feature signatures (DIFS).

arXiv:2003.08821

DHOG: Deep Hierarchical Object Grouping

Recently, a number of competitive methods have tackled unsupervised representation learning by maximising the mutual information between the representations produced from augmentations. The resulting representations are then invariant to stochastic augmentation strategies, and can be used for downstream tasks such as clustering or classification. Yet data augmentations preserve many properties of an image and so there is potential for a suboptimal choice of representation that relies on matching easy-to-find features in the data. We demonstrate that greedy or local methods of maximising mutual information (such as stochastic gradient optimisation) discover local optima of the mutual information criterion; the resulting representations are also less-ideally suited to complex downstream tasks. Earlier work has not specifically identified or addressed this issue. We introduce deep hierarchical object grouping (DHOG) that computes a number of distinct discrete representations of images in a hierarchical order, eventually generating representations that better optimise the mutual information objective. We also find that these representations align better with the downstream task of grouping into underlying object classes. We tested DHOG on unsupervised clustering, which is a natural downstream test as the target representation is a discrete labelling of the data. We achieved new state-of-the-art results on the three main benchmarks without any prefiltering or Sobel-edge detection that proved necessary for many previous methods to work. We obtain accuracy improvements of: 4.3% on CIFAR-10, 1.5% on CIFAR-100-20, and 7.2% on SVHN.

arXiv:2003.08803

Deep Object Detection based Mitosis Analysis in Breast Cancer Histopathological Images

Empirical evaluation of breast tissue biopsies for mitotic nuclei detection is considered an important prognostic biomarker in tumor grading and cancer progression. However, automated mitotic nuclei detection poses several challenges because of the unavailability of pixel-level annotations, different morphological configurations of mitotic nuclei, their sparse representation, and close resemblance with non-mitotic nuclei. These challenges undermine the precision of the automated detection model and thus make detection difficult in a single phase. This work proposes an end-to-end detection system for mitotic nuclei identification in breast cancer histopathological images. Deep object detection-based Mask R-CNN is adapted for mitotic nuclei detection that initially selects the candidate mitotic region with maximum recall. However, in the second phase, these candidate regions are refined by multi-object loss function to improve the precision. The performance of the proposed detection model shows improved discrimination ability (F-score of 0.86) for mitotic nuclei with significant precision (0.86) as compared to the two-stage detection models (F-score of 0.701) on TUPAC16 dataset. Promising results suggest that the deep object detection-based model has the potential to learn the characteristic features of mitotic nuclei from weakly annotated data and suggests that it can be adapted for the identification of other nuclear bodies in histopathological images.

arXiv:2003.08798공식 코드 · Papers with Code

Incremental Object Detection via Meta-Learning

In a real-world setting, object instances from new classes can be continuously encountered by object detectors. When existing object detectors are applied to such scenarios, their performance on old classes deteriorates significantly. A few efforts have been reported to address this limitation, all of which apply variants of knowledge distillation to avoid catastrophic forgetting. We note that although distillation helps to retain previous learning, it obstructs fast adaptability to new tasks, which is a critical requirement for incremental learning. In this pursuit, we propose a meta-learning approach that learns to reshape model gradients, such that information across incremental tasks is optimally shared. This ensures a seamless information transfer via a meta-learned gradient preconditioning that minimizes forgetting and maximizes knowledge transfer. In comparison to existing meta-learning methods, our approach is task-agnostic, allows incremental addition of new-classes and scales to high-capacity models for object detection. We evaluate our approach on a variety of incremental learning settings defined on PASCAL-VOC and MS COCO datasets, where our approach performs favourably well against state-of-the-art methods.

arXiv:2003.08793

Deep Active Learning for Remote Sensing Object Detection

Recently, CNN object detectors have achieved high accuracy on remote sensing images but require huge labor and time costs on annotation. In this paper, we propose a new uncertainty-based active learning which can select images with more information for annotation and detector can still reach high performance with a fraction of the training images. Our method not only analyzes objects' classification uncertainty to find least confident objects but also considers their regression uncertainty to declare outliers. Besides, we bring out two extra weights to overcome two difficulties in remote sensing datasets, class-imbalance and difference in images' objects amount. We experiment our active learning algorithm on DOTA dataset with CenterNet as object detector. We achieve same-level performance as full supervision with only half images. We even override full supervision with 55% images and augmented weights on least confident images.

arXiv:2003.08603

HyNNA: Improved Performance for Neuromorphic Vision Sensor based Surveillance using Hybrid Neural Network Architecture

Applications in the Internet of Video Things (IoVT) domain have very tight constraints with respect to power and area. While neuromorphic vision sensors (NVS) may offer advantages over traditional imagers in this domain, the existing NVS systems either do not meet the power constraints or have not demonstrated end-to-end system performance. To address this, we improve on a recently proposed hybrid event-frame approach by using morphological image processing algorithms for region proposal and address the low-power requirement for object detection and classification by exploring various convolutional neural network (CNN) architectures. Specifically, we compare the results obtained from our object detection framework against the state-of-the-art low-power NVS surveillance system and show an improved accuracy of 82.16% from 63.1%. Moreover, we show that using multiple bits does not improve accuracy, and thus, system designers can save power and area by using only single bit event polarity information. In addition, we explore the CNN architecture space for object classification and show useful insights to trade-off accuracy for lower power using lesser memory and arithmetic operations.

arXiv:2003.08526공식 코드 · Papers with Code

Pose Augmentation: Class-agnostic Object Pose Transformation for Object Recognition

Object pose increases intraclass object variance which makes object recognition from 2D images harder. To render a classifier robust to pose variations, most deep neural networks try to eliminate the influence of pose by using large datasets with many poses for each class. Here, we propose a different approach: a class-agnostic object pose transformation network (OPT-Net) can transform an image along 3D yaw and pitch axes to synthesize additional poses continuously. Synthesized images lead to better training of an object classifier. We design a novel eliminate-add structure to explicitly disentangle pose from object identity: first eliminate pose information of the input image and then add target pose information (regularized as continuous variables) to synthesize any target pose. We trained OPT-Net on images of toy vehicles shot on a turntable from the iLab-20M dataset. After training on unbalanced discrete poses (5 classes with 6 poses per object instance, plus 5 classes with only 2 poses), we show that OPT-Net can synthesize balanced continuous new poses along yaw and pitch axes with high quality. Training a ResNet-18 classifier with original plus synthesized poses improves mAP accuracy by 9% overtraining on original poses only. Further, the pre-trained OPT-Net can generalize to new object classes, which we demonstrate on both iLab-20M and RGB-D. We also show that the learned features can generalize to ImageNet.

arXiv:2003.08514공식 코드 · Papers with Code

Evaluating Salient Object Detection in Natural Images with Multiple Objects having Multi-level Saliency

Salient object detection is evaluated using binary ground truth with the labels being salient object class and background. In this paper, we corroborate based on three subjective experiments on a novel image dataset that objects in natural images are inherently perceived to have varying levels of importance. Our dataset, named SalMoN (saliency in multi-object natural images), has 588 images containing multiple objects. The subjective experiments performed record spontaneous attention and perception through eye fixation duration, point clicking and rectangle drawing. As object saliency in a multi-object image is inherently multi-level, we propose that salient object detection must be evaluated for the capability to detect all multi-level salient objects apart from the salient object class detection capability. For this purpose, we generate multi-level maps as ground truth corresponding to all the dataset images using the results of the subjective experiments, with the labels being multi-level salient objects and background. We then propose the use of mean absolute error, Kendall's rank correlation and average area under precision-recall curve to evaluate existing salient object detection methods on our multi-level saliency ground truth dataset. Approaches that represent saliency detection on images as local-global hierarchical processing of a graph perform well in our dataset.

arXiv:2003.08717

Giving Commands to a Self-driving Car: A Multimodal Reasoner for Visual Grounding

We propose a new spatial memory module and a spatial reasoner for the Visual Grounding (VG) task. The goal of this task is to find a certain object in an image based on a given textual query. Our work focuses on integrating the regions of a Region Proposal Network (RPN) into a new multi-step reasoning model which we have named a Multimodal Spatial Region Reasoner (MSRR). The introduced model uses the object regions from an RPN as initialization of a 2D spatial memory and then implements a multi-step reasoning process scoring each region according to the query, hence why we call it a multimodal reasoner. We evaluate this new model on challenging datasets and our experiments show that our model that jointly reasons over the object regions of the image and words of the query largely improves accuracy compared to current state-of-the-art models.

arXiv:2003.08033

Object-Based Image Coding: A Learning-Driven Revisit

The Object-Based Image Coding (OBIC) that was extensively studied about two decades ago, promised a vast application perspective for both ultra-low bitrate communication and high-level semantical content understanding, but it had rarely been used due to the inefficient compact representation of object with arbitrary shape. A fundamental issue behind is how to efficiently process the arbitrary-shaped objects at a fine granularity (e.g., feature element or pixel wise). To attack this, we have proposed to apply the element-wise masking and compression by devising an object segmentation network for image layer decomposition, and parallel convolution-based neural image compression networks to process masked foreground objects and background scene separately. All components are optimized in an end-to-end learning framework to intelligently weigh their (e.g., object and background) contributions for visually pleasant reconstruction. We have conducted comprehensive experiments to evaluate the performance on PASCAL VOC dataset at a very low bitrate scenario (e.g., $\lesssim$0.1 bits per pixel - bpp) which have demonstrated noticeable subjective quality improvement compared with JPEG2K, HEVC-based BPG and another learned image compression method. All relevant materials are made publicly accessible at https://njuvision.github.io/Neural-Object-Coding/.

arXiv:2003.08021

Applying r-spatiogram in object tracking for occlusion handling

Object tracking is one of the most important problems in computer vision. The aim of video tracking is to extract the trajectories of a target or object of interest, i.e. accurately locate a moving target in a video sequence and discriminate target from non-targets in the feature space of the sequence. So, feature descriptors can have significant effects on such discrimination. In this paper, we use the basic idea of many trackers which consists of three main components of the reference model, i.e., object modeling, object detection and localization, and model updating. However, there are major improvements in our system. Our forth component, occlusion handling, utilizes the r-spatiogram to detect the best target candidate. While spatiogram contains some moments upon the coordinates of the pixels, r-spatiogram computes region-based compactness on the distribution of the given feature in the image that captures richer features to represent the objects. The proposed research develops an efficient and robust way to keep tracking the object throughout video sequences in the presence of significant appearance variations and severe occlusions. The proposed method is evaluated on the Princeton RGBD tracking dataset considering sequences with different challenges and the obtained results demonstrate the effectiveness of the proposed method.

arXiv:2003.08333공식 코드 · Papers with Code

Collaborative Video Object Segmentation by Foreground-Background Integration

This paper investigates the principles of embedding learning to tackle the challenging semi-supervised video object segmentation. Different from previous practices that only explore the embedding learning using pixels from foreground object (s), we consider background should be equally treated and thus propose Collaborative video object segmentation by Foreground-Background Integration (CFBI) approach. Our CFBI implicitly imposes the feature embedding from the target foreground object and its corresponding background to be contrastive, promoting the segmentation results accordingly. With the feature embedding from both foreground and background, our CFBI performs the matching process between the reference and the predicted sequence from both pixel and instance levels, making the CFBI be robust to various object scales. We conduct extensive experiments on three popular benchmarks, i.e., DAVIS 2016, DAVIS 2017, and YouTube-VOS. Our CFBI achieves the performance (J$F) of 89.4%, 81.9%, and 81.4%, respectively, outperforming all the other state-of-the-art methods. Code: https://github.com/z-x-yang/CFBI.

arXiv:2003.08230공식 코드 · Papers with Code

Rethinking Object Detection in Retail Stores

The convention standard for object detection uses a bounding box to represent each individual object instance. However, it is not practical in the industry-relevant applications in the context of warehouses due to severe occlusions among groups of instances of the same categories. In this paper, we propose a new task, ie, simultaneously object localization and counting, abbreviated as Locount, which requires algorithms to localize groups of objects of interest with the number of instances. However, there does not exist a dataset or benchmark designed for such a task. To this end, we collect a large-scale object localization and counting dataset with rich annotations in retail stores, which consists of 50,394 images with more than 1.9 million object instances in 140 categories. Together with this dataset, we provide a new evaluation protocol and divide the training and testing subsets to fairly evaluate the performance of algorithms for Locount, developing a new benchmark for the Locount task. Moreover, we present a cascaded localization and counting network as a strong baseline, which gradually classifies and regresses the bounding boxes of objects with the predicted numbers of instances enclosed in the bounding boxes, trained in an end-to-end manner. Extensive experiments are conducted on the proposed dataset to demonstrate its significance and the analysis discussions on failure cases are provided to indicate future directions. Dataset is available at https://isrc.iscas.ac.cn/gitlab/research/locount-dataset.

arXiv:2003.08151

The State of Lifelong Learning in Service Robots: Current Bottlenecks in Object Perception and Manipulation

Service robots are appearing more and more in our daily life. The development of service robots combines multiple fields of research, from object perception to object manipulation. The state-of-the-art continues to improve to make a proper coupling between object perception and manipulation. This coupling is necessary for service robots not only to perform various tasks in a reasonable amount of time but also to continually adapt to new environments and safely interact with non-expert human users. Nowadays, robots are able to recognize various objects, and quickly plan a collision-free trajectory to grasp a target object in predefined settings. Besides, in most of the cases, there is a reliance on large amounts of training data. Therefore, the knowledge of such robots is fixed after the training phase, and any changes in the environment require complicated, time-consuming, and expensive robot re-programming by human experts. Therefore, these approaches are still too rigid for real-life applications in unstructured environments, where a significant portion of the environment is unknown and cannot be directly sensed or controlled. In such environments, no matter how extensive the training data used for batch learning, a robot will always face new objects. Therefore, apart from batch learning, the robot should be able to continually learn about new object categories and grasp affordances from very few training examples on-site. Moreover, apart from robot self-learning, non-expert users could interactively guide the process of experience acquisition by teaching new concepts, or by correcting insufficient or erroneous concepts. In this way, the robot will constantly learn how to help humans in everyday tasks by gaining more and more experiences without the need for re-programming.

arXiv:2003.07685공식 코드 · Papers with Code

Weakly-Supervised Salient Object Detection via Scribble Annotations

Compared with laborious pixel-wise dense labeling, it is much easier to label data by scribbles, which only costs 1$\sim$2 seconds to label one image. However, using scribble labels to learn salient object detection has not been explored. In this paper, we propose a weakly-supervised salient object detection model to learn saliency from such annotations. In doing so, we first relabel an existing large-scale salient object detection dataset with scribbles, namely S-DUTS dataset. Since object structure and detail information is not identified by scribbles, directly training with scribble labels will lead to saliency maps of poor boundary localization. To mitigate this problem, we propose an auxiliary edge detection task to localize object edges explicitly, and a gated structure-aware loss to place constraints on the scope of structure to be recovered. Moreover, we design a scribble boosting scheme to iteratively consolidate our scribble annotations, which are then employed as supervision to learn high-quality saliency maps. As existing saliency evaluation metrics neglect to measure structure alignment of the predictions, the saliency map ranking metric may not comply with human perception. We present a new metric, termed saliency structure measure, to measure the structure alignment of the predicted saliency maps, which is more consistent with human perception. Extensive experiments on six benchmark datasets demonstrate that our method not only outperforms existing weakly-supervised/unsupervised methods, but also is on par with several fully-supervised state-of-the-art models. Our code and data is publicly available at https://github.com/JingZhang617/Scribble_Saliency.

arXiv:2003.07557

1st Place Solutions for OpenImage2019 -- Object Detection and Instance Segmentation

This article introduces the solutions of the two champion teams, `MMfruit' for the detection track and `MMfruitSeg' for the segmentation track, in OpenImage Challenge 2019. It is commonly known that for an object detector, the shared feature at the end of the backbone is not appropriate for both classification and regression, which greatly limits the performance of both single stage detector and Faster RCNN \cite{ren2015faster} based detector. In this competition, we observe that even with a shared feature, different locations in one object has completely inconsistent performances for the two tasks. \textit{E.g. the features of salient locations are usually good for classification, while those around the object edge are good for regression.} Inspired by this, we propose the Decoupling Head (DH) to disentangle the object classification and regression via the self-learned optimal feature extraction, which leads to a great improvement. Furthermore, we adjust the soft-NMS algorithm to adj-NMS to obtain stable performance improvement. Finally, a well-designed ensemble strategy via voting the bounding box location and confidence is proposed. We will also introduce several training/inferencing strategies and a bag of tricks that give minor improvement. Given those masses of details, we train and aggregate 28 global models with various backbones, heads and 3+2 expert models, and achieves the 1st place on the OpenImage 2019 Object Detection Challenge on the both public and private leadboards. Given such good instance bounding box, we further design a simple instance-level semantic segmentation pipeline and achieve the 1st place on the segmentation challenge.

arXiv:2003.07540

Revisiting the Sibling Head in Object Detector

The ``shared head for classification and localization'' (sibling head), firstly denominated in Fast RCNN~\cite{girshick2015fast}, has been leading the fashion of the object detection community in the past five years. This paper provides the observation that the spatial misalignment between the two object functions in the sibling head can considerably hurt the training process, but this misalignment can be resolved by a very simple operator called task-aware spatial disentanglement (TSD). Considering the classification and regression, TSD decouples them from the spatial dimension by generating two disentangled proposals for them, which are estimated by the shared proposal. This is inspired by the natural insight that for one instance, the features in some salient area may have rich information for classification while these around the boundary may be good at bounding box regression. Surprisingly, this simple design can boost all backbones and models on both MS COCO and Google OpenImage consistently by ~3% mAP. Further, we propose a progressive constraint to enlarge the performance margin between the disentangled and the shared proposals, and gain ~1% more mAP. We show the \algname{} breaks through the upper bound of nowadays single-model detector by a large margin (mAP 49.4 with ResNet-101, 51.2 with SENet154), and is the core model of our 1st place solution on the Google OpenImage Challenge 2019.

arXiv:2003.07525

Cooperative Object Detection and Parameter Estimation Using Visible Light Communications

Visible light communication (VLC) systems are promising candidates for future indoor access and peer-to-peer networks. The performance of these systems, however, is vulnerable to the line of sight (LOS) link blockage due to objects inside the room. In this paper, we develop a probabilistic object detection method that takes advantage of the blockage status of the LOS links between the user devices and transceivers on the ceiling to locate those objects. The target objects are modeled as cylinders with random radii. The location and size of an object can be estimated by using a quadratic programming approach. Simulation results show that the root-mean-squared error can be less than $1$ cm and $8$ cm for estimating the center and the radius of the object, respectively.

arXiv:2003.07449

Object-Centric Image Generation from Layouts

Despite recent impressive results on single-object and single-domain image generation, the generation of complex scenes with multiple objects remains challenging. In this paper, we start with the idea that a model must be able to understand individual objects and relationships between objects in order to generate complex scenes well. Our layout-to-image-generation method, which we call Object-Centric Generative Adversarial Network (or OC-GAN), relies on a novel Scene-Graph Similarity Module (SGSM). The SGSM learns representations of the spatial relationships between objects in the scene, which lead to our model's improved layout-fidelity. We also propose changes to the conditioning mechanism of the generator that enhance its object instance-awareness. Apart from improving image quality, our contributions mitigate two failure modes in previous approaches: (1) spurious objects being generated without corresponding bounding boxes in the layout, and (2) overlapping bounding boxes in the layout leading to merged objects in images. Extensive quantitative evaluation and ablation studies demonstrate the impact of our contributions, with our model outperforming previous state-of-the-art approaches on both the COCO-Stuff and Visual Genome datasets. Finally, we address an important limitation of evaluation metrics used in previous works by introducing SceneFID -- an object-centric adaptation of the popular Fr{\'e}chet Inception Distance metric, that is better suited for multi-object images.