Exploiting multi-scale representations is critical to improve edge detection for objects at different scales. To extract edges dramatically scales, we propose a Bi-Directional Cascade Network (BDCN) structure, where an individual layer supervised by labeled its specific scale, rather than directly applying the same supervision all CNN outputs. Furthermore, enrich learned BDCN, introduce Scale Enhancement Module (SEM) which utilizes dilated convolution generate features, instead of using...

Recently, proposal-free instance segmentation has received increasing attention due to its concise and efficient pipeline. Generally, methods generate instance-agnostic semantic labels instance-aware features group pixels into different object instances. However, previous mostly employ separate modules for these two sub-tasks require multiple passes inference. We argue that treating separately is suboptimal. In fact, employing significantly reduces the potential application. The mutual...

Exploiting multi-scale representations is critical to improve edge detection for objects at different scales. To extract edges dramatically scales, we propose a bi-directional cascade network (BDCN) architecture, where an individual layer supervised by labeled its specific scale, rather than directly applying the same supervision layers. Furthermore, enrich learned each of BDCN, introduce scale enhancement module (SEM), which utilizes dilated convolution generate features, instead using...

Proposal-free instance segmentation methods mainly generate instance-agnostic semantic labels and instance-aware features to group pixels into different object instances. However, previous mostly employ separate modules for these two sub-tasks require multiple passes inference. In addition the lack of efficiency, also failed perform as well proposal-based approaches. To this end, work proposes a single-shot proposal-free method that requires only one single pass prediction. Our is based on...

Continuum manipulators with structural compliance can be utilized to steer a laser beam in constrained environments. However, the flexibility and nonlinear characteristics of continuum bring difficulty precision manipulation. This study proposes model-free control approach visual feedback tendon-driven flexible manipulator that integrated into an endoscope accurately. To overcome noise from disturbances environment during operation, local Jacobian matrix maps actuation space image is...

Image contrast enhancement uses the object intensity transformation function to maximize amount of information enhance an image. In this paper, image problem is regarded as optimization problem, and particle swarm algorithm used obtain optimal solution. First, improved proposed. algorithm, individual optimization, local global are adjust particle’s flight direction. topology induce comparison communication between particles. The sparse penalty term in speed update formula added sparsity size...

Panoptic segmentation (PS) is a complex scene understanding task that requires providing high-quality for both thing objects and stuff regions. Previous methods handle these two classes with semantic instance modules separately, following heuristic fusion or additional to resolve the conflicts between outputs. This work simplifies this pipeline of PS by consistently modeling novel framework, which extends detection model an extra module predict category- instance-aware pixel embedding...

This paper presents a unified framework for depth-aware panoptic segmentation (DPS), which aims to reconstruct 3D scene with instance-level semantics from one single image. Prior works address this problem by simply adding dense depth regression head (PS) networks, resulting in two independent task branches. neglects the mutually-beneficial relations between these tasks, thus failing exploit handy semantic cues boost accuracy while also producing sub-optimal maps. To overcome limitations, we...

Exploiting multi-scale representations is critical to improve edge detection for objects at different scales. To extract edges dramatically scales, we propose a Bi-Directional Cascade Network (BDCN) structure, where an individual layer supervised by labeled its specific scale, rather than directly applying the same supervision all CNN outputs. Furthermore, enrich learned BDCN, introduce Scale Enhancement Module (SEM) which utilizes dilated convolution generate features, instead of using...

Recently, proposal-free instance segmentation has received increasing attention due to its concise and efficient pipeline. Generally, methods generate instance-agnostic semantic labels instance-aware features group pixels into different object instances. However, previous mostly employ separate modules for these two sub-tasks require multiple passes inference. We argue that treating separately is suboptimal. In fact, employing significantly reduces the potential application. The mutual...

Video instance segmentation (VIS) aims at segmenting and tracking objects in videos. Prior methods typically generate frame-level or clip-level object instances first then associate them by either additional heads complex matching algorithms. This explicit association approach increases system complexity fails to fully exploit temporal cues In this paper, we design a simple, fast yet effective query-based framework for online VIS. Relying on an query proposal propagation mechanism with...

Common action recognition methods describe an sequence along with its time axis, i.e., first extracting features from the x y plane, and then modeling dynamic changes axis. Other than ordinary plane-based representation, other views, e.g., xt slice-based may be more efficient to distinguish different actions. In this paper, we investigate slicing views of spatiotemporal volume organize sequences propose slice representation for human recognition. First, a minimum average entropy principle is...

Abstract The telemetry data are essential in evaluating the performance of aircraft and diagnosing its failures. This work combines oversampling technology with run-length encoding compression algorithm an error factor to further enhance a multichannel acquisition system. Compression is carried out use FPGAs. In experiments there used pulse signals vibration signals. proposed method compared two existing methods. experimental results indicate that ratio, precision, distortion degree improved...

Estimating the precision of a multichannel telemetry data system accurately and efficiently is important because high cost flight experiments complexity data. System significantly determined by crosstalk effect. This paper proposes an efficient estimation method for acquisition based on phase-shift rectangular waveforms. The are conducted to test system. Experimental results show that proposed approach can estimate accurately; this finding was further validated comparing with those IEEE Std...

Low response rate limits the effective application of immunotherapy, in which interactions between tumor cells and immune play a significant role. The strength regulation could be mediated by extracellular matrix (ECM) fibers, is still insufficiently investigated. In study, cellular potts model was utilized to explore role morphological properties ECM tumor-immune interactions. It observed that high-density random fibers delayed interaction T cells. Moreover, were morphology-specific. Radial...

Spatio-Temporal Interest Point (STIP) has been widely used for human action recognition. However, the performance of STIP based methods are still limited in realistic datasets which often include large variations illuminations, viewpoints and camera motions. One reason low is that STIPs only reflect local change videos, not enough to obtain stable informative features representation scene. To tackle problem, we proposed an approach selecting "stable STIPs" with spatio-temporal distribution...