Most existing approaches to video instance segmentation comprise multiple modules that are heuristically combined to produce the final output. Formulating a purely learning-based method instead, which models both the temporal aspect as well as a generic track management required to solve the video instance segmentation task, is a highly challenging problem. In this work, we propose a novel learning formulation, where the entire video instance segmentation problem is modelled jointly. We fit a flexible model to our formulation that, with the help of a graph neural network, processes all available new information in each frame. Past information is considered and processed via a recurrent connection. We demonstrate the effectiveness of the proposed approach in comprehensive experiments. Our approach, operating at over 25 FPS, outperforms previous video real-time methods. We further conduct detailed ablative experiments that validate the different aspects of our approach.
A person is usually characterized by descriptors like age, gender, height, cloth type, pattern, color, etc. Such descriptors are known as attributes and/or soft-biometrics. They link the semantic gap between a person's description and retrieval in video surveillance. Retrieving a specific person with the query of semantic description has an important application in video surveillance. Using computer vision to fully automate the person retrieval task has been gathering interest within the research community. However, the Current, trend mainly focuses on retrieving persons with image-based queries, which have major limitations for practical usage. Instead of using an image query, in this paper, we study the problem of person retrieval in video surveillance with a semantic description. To solve this problem, we develop a deep learning-based cascade filtering approach (PeR-ViS), which uses Mask R-CNN [14] (person detection and instance segmentation) and DenseNet-161 [16] (soft-biometric classification). On the standard person retrieval dataset of SoftBioSearch [6], we achieve 0.566 Average IoU and 0.792 %w $IoU > 0.4$, surpassing the current state-of-the-art by a large margin. We hope our simple, reproducible, and effective approach will help ease future research in the domain of person retrieval in video surveillance. The source code and pretrained weights available at https://parshwa1999.github.io/PeR-ViS/.
We present a high-performance method that can achieve mask-level instance segmentation with only bounding-box annotations for training. While this setting has been studied in the literature, here we show significantly stronger performance with a simple design (e.g., dramatically improving previous best reported mask AP of 21.1% in Hsu et al. (2019) to 31.6% on the COCO dataset). Our core idea is to redesign the loss of learning masks in instance segmentation, with no modification to the segmentation network itself. The new loss functions can supervise the mask training without relying on mask annotations. This is made possible with two loss terms, namely, 1) a surrogate term that minimizes the discrepancy between the projections of the ground-truth box and the predicted mask; 2) a pairwise loss that can exploit the prior that proximal pixels with similar colors are very likely to have the same category label. Experiments demonstrate that the redesigned mask loss can yield surprisingly high-quality instance masks with only box annotations. For example, without using any mask annotations, with a ResNet-101 backbone and 3x training schedule, we achieve 33.2% mask AP on COCO test-dev split (vs. 39.1% of the fully supervised counterpart). Our excellent experiment results on COCO and Pascal VOC indicate that our method dramatically narrows the performance gap between weakly and fully supervised instance segmentation. Code is available at: https://git.io/AdelaiDet
Panoptic segmentation, which is a novel task of unifying instance segmentation and semantic segmentation, has attracted a lot of attention lately. However, most of the previous methods are composed of multiple pathways with each pathway specialized to a designated segmentation task. In this paper, we propose to resolve panoptic segmentation in single-shot by integrating the execution flows. With the integrated pathway, a unified feature map called Panoptic-Feature is generated, which includes the information of both things and stuffs. Panoptic-Feature becomes more sophisticated by auxiliary problems that guide to cluster pixels that belong to the same instance and differentiate between objects of different classes. A collection of convolutional filters, where each filter represents either a thing or stuff, is applied to Panoptic-Feature at once, materializing the single-shot panoptic segmentation. Taking the advantages of both top-down and bottom-up approaches, our method, named SPINet, enjoys high efficiency and accuracy on major panoptic segmentation benchmarks: COCO and Cityscapes.
Object detection is a comprehensively studied problem in autonomous driving. However, it has been relatively less explored in the case of fisheye cameras. The standard bounding box fails in fisheye cameras due to the strong radial distortion, particularly in the image's periphery. We explore better representations like oriented bounding box, ellipse, and generic polygon for object detection in fisheye images in this work. We use the IoU metric to compare these representations using accurate instance segmentation ground truth. We design a novel curved bounding box model that has optimal properties for fisheye distortion models. We also design a curvature adaptive perimeter sampling method for obtaining polygon vertices, improving relative mAP score by 4.9% compared to uniform sampling. Overall, the proposed polygon model improves mIoU relative accuracy by 40.3%. It is the first detailed study on object detection on fisheye cameras for autonomous driving scenarios to the best of our knowledge. The dataset comprising of 10,000 images along with all the object representations ground truth will be made public to encourage further research. We summarize our work in a short video with qualitative results at https://youtu.be/iLkOzvJpL-A.
Analyzing complex scenes with Deep Neural Networks is a challenging task, particularly when images contain multiple objects that partially occlude each other. Existing approaches to image analysis mostly process objects independently and do not take into account the relative occlusion of nearby objects. In this paper, we propose a deep network for multi-object instance segmentation that is robust to occlusion and can be trained from bounding box supervision only. Our work builds on Compositional Networks, which learn a generative model of neural feature activations to locate occluders and to classify objects based on their non-occluded parts. We extend their generative model to include multiple objects and introduce a framework for efficient inference in challenging occlusion scenarios. In particular, we obtain feed-forward predictions of the object classes and their instance and occluder segmentations. We introduce an Occlusion Reasoning Module (ORM) that locates erroneous segmentations and estimates the occlusion order to correct them. The improved segmentation masks are, in turn, integrated into the network in a top-down manner to improve the image classification. Our experiments on the KITTI INStance dataset (KINS) and a synthetic occlusion dataset demonstrate the effectiveness and robustness of our model at multi-object instance segmentation under occlusion. Code is publically available at https://github.com/XD7479/Multi-Object-Occlusion.
Recent advances in 3D semantic scene understanding have shown impressive progress in 3D instance segmentation, enabling object-level reasoning about 3D scenes; however, a finer-grained understanding is required to enable interactions with objects and their functional understanding. Thus, we propose the task of part-based scene understanding of real-world 3D environments: from an RGB-D scan of a scene, we detect objects, and for each object predict its decomposition into geometric part masks, which composed together form the complete geometry of the observed object. We leverage an intermediary part graph representation to enable robust completion as well as building of part priors, which we use to construct the final part mask predictions. Our experiments demonstrate that guiding part understanding through part graph to part prior-based predictions significantly outperforms alternative approaches to the task of semantic part completion.
We present a novel explicit shape representation for instance segmentation. Based on how to model the object shape, current instance segmentation systems can be divided into two categories, implicit and explicit models. The implicit methods, which represent the object mask/contour by intractable network parameters, and produce it through pixel-wise classification, are predominant. However, the explicit methods, which parameterize the shape with simple and explainable models, are less explored. Since the operations to generate the final shape are light-weighted, the explicit methods have a clear speed advantage over implicit methods, which is crucial for real-world applications. The proposed USD-Seg adopts a linear model, sparse coding with dictionary, for object shapes. First, it learns a dictionary from a large collection of shape datasets, making any shape being able to be decomposed into a linear combination through the dictionary. Hence the name "Universal Shape Dictionary". Then it adds a simple shape vector regression head to ordinary object detector, giving the detector segmentation ability with minimal overhead. For quantitative evaluation, we use both average precision (AP) and the proposed Efficiency of AP (AP$_E$) metric, which intends to also measure the computational consumption of the framework to cater to the requirements of real-world applications. We report experimental results on the challenging COCO dataset, in which our single model on a single Titan Xp GPU achieves 35.8 AP and 27.8 AP$_E$ at 65 fps with YOLOv4 as base detector, 34.1 AP and 28.6 AP$_E$ at 12 fps with FCOS as base detector.
Blastomere instance segmentation is important for analyzing embryos' abnormality. To measure the accurate shapes and sizes of blastomeres, their amodal segmentation is necessary. Amodal instance segmentation aims to recover the complete silhouette of an object even when the object is not fully visible. For each detected object, previous methods directly regress the target mask from input features. However, images of an object under different amounts of occlusion should have the same amodal mask output, which makes it harder to train the regression model. To alleviate the problem, we propose to classify input features into intermediate shape codes and recover complete object shapes from them. First, we pre-train the Vector Quantized Variational Autoencoder (VQ-VAE) model to learn these discrete shape codes from ground truth amodal masks. Then, we incorporate the VQ-VAE model into the amodal instance segmentation pipeline with an additional refinement module. We also detect an occlusion map to integrate occlusion information with a backbone feature. As such, our network faithfully detects bounding boxes of amodal objects. On an internal embryo cell image benchmark, the proposed method outperforms previous state-of-the-art methods. To show generalizability, we show segmentation results on the public KINS natural image benchmark. To examine the learned shape codes and model design choices, we perform ablation studies on a synthetic dataset of simple overlaid shapes. Our method would enable accurate measurement of blastomeres in in vitro fertilization (IVF) clinics, which potentially can increase IVF success rate.
Video instance segmentation (VIS) is the task that requires simultaneously classifying, segmenting and tracking object instances of interest in video. Recent methods typically develop sophisticated pipelines to tackle this task. Here, we propose a new video instance segmentation framework built upon Transformers, termed VisTR, which views the VIS task as a direct end-to-end parallel sequence decoding/prediction problem. Given a video clip consisting of multiple image frames as input, VisTR outputs the sequence of masks for each instance in the video in order directly. At the core is a new, effective instance sequence matching and segmentation strategy, which supervises and segments instances at the sequence level as a whole. VisTR frames the instance segmentation and tracking in the same perspective of similarity learning, thus considerably simplifying the overall pipeline and is significantly different from existing approaches. Without bells and whistles, VisTR achieves the highest speed among all existing VIS models, and achieves the best result among methods using single model on the YouTube-VIS dataset. For the first time, we demonstrate a much simpler and faster video instance segmentation framework built upon Transformers, achieving competitive accuracy. We hope that VisTR can motivate future research for more video understanding tasks.
Previous top-performing approaches for point cloud instance segmentation involve a bottom-up strategy, which often includes inefficient operations or complex pipelines, such as grouping over-segmented components, introducing additional steps for refining, or designing complicated loss functions. The inevitable variation in the instance scales can lead bottom-up methods to become particularly sensitive to hyper-parameter values. To this end, we propose instead a dynamic, proposal-free, data-driven approach that generates the appropriate convolution kernels to apply in response to the nature of the instances. To make the kernels discriminative, we explore a large context by gathering homogeneous points that share identical semantic categories and have close votes for the geometric centroids. Instances are then decoded by several simple convolutional layers. Due to the limited receptive field introduced by the sparse convolution, a small light-weight transformer is also devised to capture the long-range dependencies and high-level interactions among point samples. The proposed method achieves promising results on both ScanetNetV2 and S3DIS, and this performance is robust to the particular hyper-parameter values chosen. It also improves inference speed by more than 25% over the current state-of-the-art. Code is available at: https://git.io/DyCo3D
Pursuing a more coherent scene understanding towards real-time vision applications, single-stage instance segmentation has recently gained popularity, achieving a simpler and more efficient design than its two-stage counterparts. Besides, its global mask representation often leads to superior accuracy to the two-stage Mask R-CNN which has been dominant thus far. Despite the promising advances in single-stage methods, finer delineation of instance boundaries still remains unexcavated. Indeed, boundary information provides a strong shape representation that can operate in synergy with the fully-convolutional mask features of the single-stage segmenter. In this work, we propose Boundary Basis based Instance Segmentation(B2Inst) to learn a global boundary representation that can complement existing global-mask-based methods that are often lacking high-frequency details. Besides, we devise a unified quality measure of both mask and boundary and introduce a network block that learns to score the per-instance predictions of itself. When applied to the strongest baselines in single-stage instance segmentation, our B2Inst leads to consistent improvements and accurately parse out the instance boundaries in a scene. Regardless of being single-stage or two-stage frameworks, we outperform the existing state-of-the-art methods on the COCO dataset with the same ResNet-50 and ResNet-101 backbones.
Instance segmentation of overlapping objects in biomedical images remains a largely unsolved problem. We take up this challenge and present MultiStar, an extension to the popular instance segmentation method StarDist. The key novelty of our method is that we identify pixels at which objects overlap and use this information to improve proposal sampling and to avoid suppressing proposals of truly overlapping objects. This allows us to apply the ideas of StarDist to images with overlapping objects, while incurring only a small overhead compared to the established method. MultiStar shows promising results on two datasets and has the advantage of using a simple and easy to train network architecture.
This paper proposes a multi-view extension of instance segmentation without relying on texture or shape descriptor matching. Multi-view instance segmentation becomes challenging for scenes with repetitive textures and shapes, e.g., plant leaves, due to the difficulty of multi-view matching using texture or shape descriptors. To this end, we propose a multi-view region matching method based on epipolar geometry, which does not rely on any feature descriptors. We further show that the epipolar region matching can be easily integrated into instance segmentation and effective for instance-wise 3D reconstruction. Experiments demonstrate the improved accuracy of multi-view instance matching and the 3D reconstruction compared to the baseline methods.
We introduce CellSegmenter, a structured deep generative model and an amortized inference framework for unsupervised representation learning and instance segmentation tasks. The proposed inference algorithm is convolutional and parallelized, without any recurrent mechanisms, and is able to resolve object-object occlusion while simultaneously treating distant non-occluding objects independently. This leads to extremely fast training times while allowing extrapolation to arbitrary number of instances. We further introduce a transparent posterior regularization strategy that encourages scene reconstructions with fewest localized objects and a low-complexity background. We evaluate our method on a challenging synthetic multi-MNIST dataset with a structured background and achieve nearly perfect accuracy with only a few hundred training epochs. Finally, we show segmentation results obtained for a cell nuclei imaging dataset, demonstrating the ability of our method to provide high-quality segmentations while also handling realistic use cases involving large number of instances.
While knowledge distillation (transfer) has been attracting attentions from the research community, the recent development in the fields has heightened the need for reproducible studies and highly generalized frameworks to lower barriers to such high-quality, reproducible deep learning research. Several researchers voluntarily published frameworks used in their knowledge distillation studies to help other interested researchers reproduce their original work. Such frameworks, however, are usually neither well generalized nor maintained, thus researchers are still required to write a lot of code to refactor/build on the frameworks for introducing new methods, models, datasets and designing experiments. In this paper, we present our developed open-source framework built on PyTorch and dedicated for knowledge distillation studies. The framework is designed to enable users to design experiments by declarative PyYAML configuration files, and helps researchers complete the recently proposed ML Code Completeness Checklist. Using the developed framework, we demonstrate its various efficient training strategies, and implement a variety of knowledge distillation methods. We also reproduce some of their original experimental results on the ImageNet and COCO datasets presented at major machine learning conferences such as ICLR, NeurIPS, CVPR and ECCV, including recent state-of-the-art methods. All the source code, configurations, log files and trained model weights are publicly available at https://github.com/yoshitomo-matsubara/torchdistill .
Instance segmentation methods require large datasets with expensive and thus limited instance-level mask labels. Partially supervised instance segmentation aims to improve mask prediction with limited mask labels by utilizing the more abundant weak box labels. In this work, we show that a class agnostic mask head, commonly used in partially supervised instance segmentation, has difficulties learning a general concept of foreground for the weakly annotated classes using box supervision only. To resolve this problem we introduce an object mask prior (OMP) that provides the mask head with the general concept of foreground implicitly learned by the box classification head under the supervision of all classes. This helps the class agnostic mask head to focus on the primary object in a region of interest (RoI) and improves generalization to the weakly annotated classes. We test our approach on the COCO dataset using different splits of strongly and weakly supervised classes. Our approach significantly improves over the Mask R-CNN baseline and obtains competitive performance with the state-of-the-art, while offering a much simpler architecture.
The Wide Residual Networks (Wide-ResNets), a shallow but wide model variant of the Residual Networks (ResNets) by stacking a small number of residual blocks with large channel sizes, have demonstrated outstanding performance on multiple dense prediction tasks. However, since proposed, the Wide-ResNet architecture has barely evolved over the years. In this work, we revisit its architecture design for the recent challenging panoptic segmentation task, which aims to unify semantic segmentation and instance segmentation. A baseline model is obtained by incorporating the simple and effective Squeeze-and-Excitation and Switchable Atrous Convolution to the Wide-ResNets. Its network capacity is further scaled up or down by adjusting the width (i.e., channel size) and depth (i.e., number of layers), resulting in a family of SWideRNets (short for Scaling Wide Residual Networks). We demonstrate that such a simple scaling scheme, coupled with grid search, identifies several SWideRNets that significantly advance state-of-the-art performance on panoptic segmentation datasets in both the fast model regime and strong model regime.
While much progress has been made on the task of 3D point cloud registration, there still exists no learning-based method able to estimate the 6D pose of an object observed by a 2.5D sensor in a scene. The challenges of this scenario include the fact that most measurements are outliers depicting the object's surrounding context, and the mismatch between the complete 3D object model and its self-occluded observations. We introduce the first deep learning framework capable of effectively handling this scenario. Our method consists of an instance segmentation module followed by a pose estimation one. It allows us to perform 3D registration in a one-shot manner, without requiring an expensive iterative procedure. We further develop an on-the-fly rendering-based training strategy that is both time- and memory-efficient. Our experiments evidence the superiority of our approach over the state-of-the-art traditional and learning-based 3D registration methods.
Despite being widely used as a performance measure for visual detection tasks, Average Precision (AP) is limited in (i) reflecting localisation quality, (ii) interpretability and (iii) robustness to the design choices regarding its computation, and its applicability to outputs without confidence scores. Panoptic Quality (PQ), a measure proposed for evaluating panoptic segmentation (Kirillov et al., 2019), does not suffer from these limitations but is limited to panoptic segmentation. In this paper, we propose Localisation Recall Precision (LRP) Error as the average matching error of a visual detector computed based on both its localisation and classification qualities for a given confidence score threshold. LRP Error, initially proposed only for object detection by Oksuz et al. (2018), does not suffer from the aforementioned limitations and is applicable to all visual detection tasks. We also introduce Optimal LRP (oLRP) Error as the minimum LRP Error obtained over confidence scores to evaluate visual detectors and obtain optimal thresholds for deployment. We provide a detailed comparative analysis of LRP Error with AP and PQ, and use nearly 100 state-of-the-art visual detectors from seven visual detection tasks (i.e. object detection, keypoint detection, instance segmentation, panoptic segmentation, visual relationship detection, zero-shot detection and generalised zero-shot detection) using ten datasets to empirically show that LRP Error provides richer and more discriminative information than its counterparts. Code available at: https://github.com/kemaloksuz/LRP-Error