Publications by authors named "Kaiming He"

17 Publications

  • Page 1 of 1

Polyimide@Ketjenblack Composite: A Porous Organic Cathode for Fast Rechargeable Potassium-Ion Batteries.

Small 2020 Sep 19;16(38):e2002953. Epub 2020 Aug 19.

Institute of Physics and IMN MacroNano®, Technical University of Ilmenau, Ilmenau, 98693, Germany.

Potassium-ion batteries (PIBs) configurated by organic electrodes have been identified as a promising alternative to lithium-ion batteries. Here, a porous organic Polyimide@Ketjenblack is demonstrated in PIBs as a cathode, which exhibits excellent performance with a large reversible capacity (143 mAh g at 100 mA g ), high rate capability (125 and 105 mAh g at 1000 and 5000 mA g ), and long cycling stability (76% capacity retention at 2000 mA g over 1000 cycles). The domination of fast capacitive-like reaction kinetics is verified, which benefits from the porous structure synthesized using in situ polymerization. Moreover, a renewable and low-cost full cell is demonstrated with superior rate behavior (106 mAh g at 3200 mA g ). This work proposes a strategy to design polymer electrodes for high-performance organic PIBs.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1002/smll.202002953DOI Listing
September 2020

Potential of the glasses-free three-dimensional display system in shortening the learning curve of video-assisted endoscopic surgery: a self-controlled study.

Ann Transl Med 2019 Oct;7(20):521

Department of Thoracic Surgery, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou 510120, China.

Background: One of the largest challenges in endoscopic surgical training is adapting to a two-dimensional (2D) view. The glasses-free three-dimensional (GF-3D) display system was designed to integrate the merits of both 2D and conventional 3D (C-3D) displays, allowing surgeons to perform video-assisted endoscopic surgery under a stereoscopic view without heavy and cumbersome 3D glasses.

Methods: In this study, 15 junior thoracic surgeons were divided to test one routine and one complex task three times each via traditional high-definition 2D (HD-2D) and GF-3D to determine whether there was any advantage when using the GF-3D system to acquire endoscopic skills. The duration, numbers of stitches, and distance between every two stitches were recorded for every procedure.

Results: Seven participants were enrolled in the HD-2D group and eight participants were enrolled in the GF-3D group. All 15 participants successfully completed porcine skin continuous suture and tracheal continuous anastomosis procedures three times each. For skin continuous suture, there was no significant difference between the two groups in terms of the learning curve for speed (P=0.683) and accuracy (P=0.556). For tracheal continuous anastomosis, there was a significant difference between the two groups in terms of the learning curve for speed (P=0.001), but no significant difference was observed between the two groups in terms of the learning curve for accuracy (P=0.211).

Conclusions: In summary, both HD-2D and GF-3D display systems are efficient for routine and complex endoscopic surgery. With the help of GF-3D, surgeons can acquire new complex endoscopic skills faster than HD-2D and be free from burdensome polarized glasses. More comparative studies in a clinical setting are needed to further explore the feasibility, necessity, and economic aspects of the GF-3D display system.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.21037/atm.2019.10.01DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6861795PMC
October 2019

Is the Glasses-Free 3-Dimensional Display System More Effective for Complex Video-Assisted Thoracic Surgery? A Self-Controlled Study Ex Vivo.

Surg Innov 2019 Dec 11;26(6):712-719. Epub 2019 Jul 11.

The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.

. Considering the demerits of a high-definition 2-dimensional (HD-2D) system, with its lack of stereopsis, and a conventional 3-dimensional (C-3D) system, which results in a dimmed image, we have recently developed a glasses-free 3-dimensional (GF-3D) display system for reconstruction surgeries such as video-assisted thoracic surgery (VATS) for tracheal reconstruction. . Thoracic surgeons were invited to complete thoracoscopic continuous suture of a transected porcine trachea using the HD-2D, C-3D, and GF-3D systems on separate mornings in randomized order. The duration, numbers of stitches, and distance between every 2 stitches were recorded for every procedure. The surgeons' spontaneous eye blink rate was recorded for 5 minutes before the procedure and the last 5 minutes of the procedure. . Fifteen volunteers successfully completed the tracheal reconstruction procedures in this study. Both C-3D (0.403 ± 0.064 stitch/min, < .001) and GF-3D (0.427 ± 0.079 stitch/min, < .001) showed significant advantages in speed compared with HD-2D (0.289 ± 0.065 stitch/min). Both C-3D (2.536 ± 2.223 mm, < .001) and GF-3D (2.603 ± 2.159 mm, < .001) showed significant advantages in accuracy compared with HD-2D (3.473 ± 3.403 mm). Both HD-2D (1.240 ± 0.642, < .001) and GF-3D (1.307 ± 0.894, < .001) showed significant advantages in eye fatigue compared with C-3D (3.333 ± 1.44). . All 3 available display systems are efficient for complex VATS. With the help of stereopsis, surgeons can achieve faster operation using C-3D and GF-3D systems in a thoracoscopic simulated setting. GF-3D may be a more effective display system for VATS reconstruction in terms of speed, accuracy, and eye fatigue during operations.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1553350619853136DOI Listing
December 2019

Matrine induces apoptosis via targeting CCR7 and enhances the effect of anticancer drugs in non-small cell lung cancer in vitro.

Innate Immun 2018 10 20;24(7):394-399. Epub 2018 Sep 20.

1 Department of Thoracic and Cardiovascular Surgery, The Affiliated Hospital of Southwest Medical University, Luzhou, China.

This study mainly investigated the effects of matrine on cell apoptosis and the effects of anticancer drugs in non-small cell lung cancer (NSCLC) cell lines (A549 and LK2 cells). The results showed that matrine (≥10 μM) caused a significant inhibition on cell viability and 10 and 100 μM matrine induced cell apoptosis via influencing p53, bax, casp3, and bcl-2 expressions in A549 cells. In addition, matrine significantly down-regulated C-C chemokine receptor type 7 (CCR7) expression, and blocking the down-regulation of CCR7 by exogenous chemokine ligand 21 (CCL21) treatment alleviated matrine-caused effects of apoptosis genes in A549 cells. The results were further validated in LK2 cells that matrine regulated apoptosis gene expressions, which were reversed by CCL21 treatment. Furthermore, matrine enhances the effects of cisplatin, 5-fluorouracil, and paclitaxel in A549 cells, and the anticancer effects exhibit a dosage-dependent manner. In summary, matrine induced cell apoptosis and enhanced the effects of anticancer drugs in NSCLC cells; the mechanism might be associated with the CCR7 signal.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1753425918800555DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6830874PMC
October 2018

Focal Loss for Dense Object Detection.

IEEE Trans Pattern Anal Mach Intell 2020 Feb 23;42(2):318-327. Epub 2018 Jul 23.

The highest accuracy object detectors to date are based on a two-stage approach popularized by R-CNN, where a classifier is applied to a sparse set of candidate object locations. In contrast, one-stage detectors that are applied over a regular, dense sampling of possible object locations have the potential to be faster and simpler, but have trailed the accuracy of two-stage detectors thus far. In this paper, we investigate why this is the case. We discover that the extreme foreground-background class imbalance encountered during training of dense detectors is the central cause. We propose to address this class imbalance by reshaping the standard cross entropy loss such that it down-weights the loss assigned to well-classified examples. Our novel Focal Loss focuses training on a sparse set of hard examples and prevents the vast number of easy negatives from overwhelming the detector during training. To evaluate the effectiveness of our loss, we design and train a simple dense detector we call RetinaNet. Our results show that when trained with the focal loss, RetinaNet is able to match the speed of previous one-stage detectors while surpassing the accuracy of all existing state-of-the-art two-stage detectors. Code is at: https://github.com/facebookresearch/Detectron.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2018.2858826DOI Listing
February 2020

Mask R-CNN.

IEEE Trans Pattern Anal Mach Intell 2020 02 5;42(2):386-397. Epub 2018 Jun 5.

We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance. The method, called Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN, running at 5 fps. Moreover, Mask R-CNN is easy to generalize to other tasks, e.g., allowing us to estimate human poses in the same framework. We show top results in all three tracks of the COCO suite of challenges, including instance segmentation, bounding-box object detection, and person keypoint detection. Without bells and whistles, Mask R-CNN outperforms all existing, single-model entries on every task, including the COCO 2016 challenge winners. We hope our simple and effective approach will serve as a solid baseline and help ease future research in instance-level recognition. Code has been made available at: https://github.com/facebookresearch/Detectron.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2018.2844175DOI Listing
February 2020

Object Detection Networks on Convolutional Feature Maps.

IEEE Trans Pattern Anal Mach Intell 2017 07 17;39(7):1476-1481. Epub 2016 Aug 17.

Most object detectors contain two important components: a feature extractor and an object classifier. The feature extractor has rapidly evolved with significant research efforts leading to better deep convolutional architectures. The object classifier, however, has not received much attention and many recent systems (like SPPnet and Fast/Faster R-CNN) use simple multi-layer perceptrons. This paper demonstrates that carefully designing deep networks for object classification is just as important. We experiment with region-wise classifier networks that use shared, region-independent convolutional features. We call them "Networks on Convolutional feature maps" (NoCs). We discover that aside from deep feature maps, a deep and convolutional per-region classifier is of particular importance for object detection, whereas latest superior image classification models (such as ResNets and GoogLeNets) do not directly lead to good detection accuracy without using such a per-region classifier. We show by experiments that despite the effective ResNets and Faster R-CNN systems, the design of NoCs is an essential element for the 1st-place winning entries in ImageNet and MS COCO challenges 2015.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2016.2601099DOI Listing
July 2017

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.

IEEE Trans Pattern Anal Mach Intell 2017 06 6;39(6):1137-1149. Epub 2016 Jun 6.

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet [1] and Fast R-CNN [2] have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features-using the recently popular terminology of neural networks with 'attention' mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model [3] , our detection system has a frame rate of 5 fps (including all steps) on a GPU, while achieving state-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2016.2577031DOI Listing
June 2017

Image Super-Resolution Using Deep Convolutional Networks.

IEEE Trans Pattern Anal Mach Intell 2016 Feb;38(2):295-307

We propose a deep learning method for single image super-resolution (SR). Our method directly learns an end-to-end mapping between the low/high-resolution images. The mapping is represented as a deep convolutional neural network (CNN) that takes the low-resolution image as the input and outputs the high-resolution one. We further show that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network. But unlike traditional methods that handle each component separately, our method jointly optimizes all layers. Our deep CNN has a lightweight structure, yet demonstrates state-of-the-art restoration quality, and achieves fast speed for practical on-line usage. We explore different network structures and parameter settings to achieve trade-offs between performance and speed. Moreover, we extend our network to cope with three color channels simultaneously, and show better overall reconstruction quality.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2015.2439281DOI Listing
February 2016

Accelerating Very Deep Convolutional Networks for Classification and Detection.

IEEE Trans Pattern Anal Mach Intell 2016 10 20;38(10):1943-55. Epub 2015 Nov 20.

This paper aims to accelerate the test-time computation of convolutional neural networks (CNNs), especially very deep CNNs [1] that have substantially impacted the computer vision community. Unlike previous methods that are designed for approximating linear filters or linear responses, our method takes the nonlinear units into account. We develop an effective solution to the resulting nonlinear optimization problem without the need of stochastic gradient descent (SGD). More importantly, while previous methods mainly focus on optimizing one or two layers, our nonlinear method enables an asymmetric reconstruction that reduces the rapidly accumulated error when multiple (e.g., ≥ 10) layers are approximated. For the widely used very deep VGG-16 model [1] , our method achieves a whole-model speedup of 4 × with merely a 0.3 percent increase of top-5 error in ImageNet classification. Our 4 × accelerated VGG-16 model also shows a graceful accuracy degradation for object detection when plugged into the Fast R-CNN detector [2] .
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2015.2502579DOI Listing
October 2016

Image Completion Approaches Using the Statistics of Similar Patches.

Authors:
Kaiming He Jian Sun

IEEE Trans Pattern Anal Mach Intell 2014 Dec;36(12):2423-35

Image completion involves filling missing parts in images. In this paper we address this problem through novel statistics of similar patches. We observe that if we match similar patches in the image and obtain their offsets (relative positions), the statistics of these offsets are sparsely distributed. We further observe that a few dominant offsets provide reliable information for completing the image. Such statistics can be incorporated into both matching-based and graph-based methods for image completion. Experiments show that our method yields better results in various challenging cases, and is faster than existing state-of-the-art methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2014.2330611DOI Listing
December 2014

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition.

IEEE Trans Pattern Anal Mach Intell 2015 Sep;37(9):1904-16

Existing deep convolutional neural networks (CNNs) require a fixed-size (e.g., 224 × 224) input image. This requirement is "artificial" and may reduce the recognition accuracy for the images or sub-images of an arbitrary size/scale. In this work, we equip the networks with another pooling strategy, "spatial pyramid pooling", to eliminate the above requirement. The new network structure, called SPP-net, can generate a fixed-length representation regardless of image size/scale. Pyramid pooling is also robust to object deformations. With these advantages, SPP-net should in general improve all CNN-based image classification methods. On the ImageNet 2012 dataset, we demonstrate that SPP-net boosts the accuracy of a variety of CNN architectures despite their different designs. On the Pascal VOC 2007 and Caltech101 datasets, SPP-net achieves state-of-the-art classification results using a single full-image representation and no fine-tuning. The power of SPP-net is also significant in object detection. Using SPP-net, we compute the feature maps from the entire image only once, and then pool features in arbitrary regions (sub-images) to generate fixed-length representations for training the detectors. This method avoids repeatedly computing the convolutional features. In processing test images, our method is 24-102 × faster than the R-CNN method, while achieving better or comparable accuracy on Pascal VOC 2007. In ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2014, our methods rank #2 in object detection and #3 in image classification among all 38 teams. This manuscript also introduces the improvement made for this competition.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2015.2389824DOI Listing
September 2015

Optimized Product Quantization.

IEEE Trans Pattern Anal Mach Intell 2014 Apr;36(4):744-55

Product quantization (PQ) is an effective vector quantization method. A product quantizer can generate an exponentially large codebook at very low memory/time cost. The essence of PQ is to decompose the high-dimensional vector space into the Cartesian product of subspaces and then quantize these subspaces separately. The optimal space decomposition is important for the PQ performance, but still remains an unaddressed issue. In this paper, we optimize PQ by minimizing quantization distortions w.r.t the space decomposition and the quantization codebooks. We present two novel solutions to this challenging optimization problem. The first solution iteratively solves two simpler sub-problems. The second solution is based on a Gaussian assumption and provides theoretical analysis of the optimality. We evaluate our optimized product quantizers in three applications: (i) compact encoding for exhaustive ranking [1], (ii) building inverted multi-indexing for non-exhaustive search [2], and (iii) compacting image representations for image retrieval [3]. In all applications our optimized product quantizers outperform existing solutions.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2013.240DOI Listing
April 2014

Guided image filtering.

IEEE Trans Pattern Anal Mach Intell 2013 Jun;35(6):1397-409

Visual Computing Group, Microsoft Research Asia, Microsoft Building 2, #5 Dan Leng Street, Hai Dian District, Beijing 100080, China.

In this paper, we propose a novel explicit image filter called guided filter. Derived from a local linear model, the guided filter computes the filtering output by considering the content of a guidance image, which can be the input image itself or another different image. The guided filter can be used as an edge-preserving smoothing operator like the popular bilateral filter [1], but it has better behaviors near edges. The guided filter is also a more generic concept beyond smoothing: It can transfer the structures of the guidance image to the filtering output, enabling new filtering applications like dehazing and guided feathering. Moreover, the guided filter naturally has a fast and nonapproximate linear time algorithm, regardless of the kernel size and the intensity range. Currently, it is one of the fastest edge-preserving filters. Experiments show that the guided filter is both effective and efficient in a great variety of computer vision and computer graphics applications, including edge-aware smoothing, detail enhancement, HDR compression, image matting/feathering, dehazing, joint upsampling, etc.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2012.213DOI Listing
June 2013

Single Image Haze Removal Using Dark Channel Prior.

IEEE Trans Pattern Anal Mach Intell 2011 Dec 9;33(12):2341-53. Epub 2010 Sep 9.

In this paper, we propose a simple but effective image prior-dark channel prior to remove haze from a single input image. The dark channel prior is a kind of statistics of outdoor haze-free images. It is based on a key observation-most local patches in outdoor haze-free images contain some pixels whose intensity is very low in at least one color channel. Using this prior with the haze imaging model, we can directly estimate the thickness of the haze and recover a high-quality haze-free image. Results on a variety of hazy images demonstrate the power of the proposed prior. Moreover, a high-quality depth map can also be obtained as a byproduct of haze removal.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TPAMI.2010.168DOI Listing
December 2011