Compared to many other dense prediction tasks, e.g., semantic segmentation, it is the arbitrary number of instances that has made instance segmentation much more challenging. In order to predict a mask for each instance, mainstream approaches either follow the "detect-then-segment" strategy (e.g., Mask R-CNN), or predict embedding vectors first then cluster pixels into individual instances. In this paper, we view the task of instance segmentation from a completely new perspective by introducing the notion of "instance categories", which assigns categories to each pixel within an instance according to the instance's location. With this notion, we propose segmenting objects by locations (SOLO), a simple, direct, and fast framework for instance segmentation with strong performance. We derive a few SOLO variants (e.g., Vanilla SOLO, Decoupled SOLO, Dynamic SOLO) following the basic principle. Our method directly maps a raw input image to the desired object categories and instance masks, eliminating the need for the grouping post-processing or the bounding box detection. Our approach achieves state-of-the-art results for instance segmentation in terms of both speed and accuracy, while being considerably simpler than the existing methods. Besides instance segmentation, our method yields state-of-the-art results in object detection (from our mask byproduct) and panoptic segmentation. We further demonstrate the flexibility and high-quality segmentation of SOLO by extending it to perform one-stage instance-level image matting.
A 3D thinning algorithm erodes a 3D binary image layer by layer to extract the skeletons. This paper presents a correction to Ma and Sonka's thinning algorithm, ' A fully parallel 3D thinning algorithm and its applications ' , which fails to preserve connectivity of 3D objects. We start with Ma and Sonka's algorithm and examine its verification of connectivity preservation. Our analysis leads to a group of different deleting templates, which can preserve connectivity of 3D objects. IntroductionThinning is a useful technique having potential applications in a wide variety of problems. It creates a compact representation (skeleton) of the models that may be used for further processing. 3D skeletons can be used in many applications [1-2] such as 3D pattern matching, 3D recognition and 3D database retrieval.A 3D binary image is a mapping that assigns the value of 0 or 1 to each point in the 3D space. Points having the value of 1 are called black (object) points, while 0's are called white (background) ones. Black points form objects of the binary image. The thinning operation iteratively deletes or removes some object points (that is, changes some black points to white) until only some restrictions prevent further operation.Note that the white points will never be changed to black ones in any circumstances. Most of the existing thinning algorithms are parallel, since the medial axis transform (MAT) can be defined as fire front propagation, which is by nature parallel [3]. There are three categories of parallel thinning algorithms in literature, sub-iteration parallel thinning algorithm [4][5][6], sub-field parallel thinning algorithm [7][8] and fully parallel thinning algorithm [9]. Brief surveys of algorithms in each category can be found in the literature [6,9].The rest of this paper is organized as follows. In Section 2, some basic concepts will be presented. Section 3 will briefly discuss Ma and Sonka's algorithm [9]. The problematic part in this algorithm is analyzed and the modification is presented in Section 4, before the work is concluded in Section 5.
We present GSPMD, an automatic, compiler-based parallelization system for common machine learning computation graphs. It allows users to write programs in the same way as for a single device, then give hints through a few annotations on how to distribute tensors, based on which GSPMD will parallelize the computation. Its representation of partitioning is simple yet general, allowing it to express different or mixed paradigms of parallelism on a wide variety of models.GSPMD infers the partitioning for every operator in the graph based on limited user annotations, making it convenient to scale up existing single-device programs. It solves several technical challenges for production usage, such as static shape constraints, uneven partitioning, exchange of halo data, and nested operator partitioning. These techniques allow GSPMD to achieve 50% to 62% compute utilization on 128 to 2048 Cloud TPUv3 cores for models with up to one trillion parameters.GSPMD produces a single program for all devices, which adjusts its behavior based on a run-time partition ID, and uses collective operators for cross-device communication. This property allows the system itself to be scalable: the compilation time stays constant with increasing number of devices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.