Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays an important role in diagnosis and grading of brain tumor. Although manual DCE biomarker extraction algorithms boost the diagnostic yield of DCE-MRI by providing quantitative information on tumor prognosis and prediction, they are time-consuming and prone to human error. In this paper, we propose a fully-automated, end-to-end system for DCE-MRI analysis of brain tumors. Our deep learning-powered technique does not require any user interaction, it yields reproducible results, and it is rigorously validated against benchmark (BraTS'17 for tumor segmentation, and a test dataset released by the Quantitative Imaging Biomarkers Alliance for the contrast-concentration fitting) and clinical (44 low-grade glioma patients) data. Also, we introduce a cubic model of the vascular input function used for pharmacokinetic modeling which significantly decreases the fitting error when compared with the state of the art, alongside a real-time algorithm for determination of the vascular input region. An extensive experimental study, backed up with statistical tests, showed that our system delivers state-of-the-art results (in terms of segmentation accuracy and contrast-concentration fitting) while requiring less than 3 minutes to process an entire input DCE-MRI study using a single GPU.
This paper contains a feasibility study of deep neural networks for the classification of Euro banknotes with respect to requirements of central banks on the ATM and high speed sorting industry. Instead of concentrating on the accuracy for a large number of classes as in the famous ImageNet Challenge we focus thus on conditions with few classes and the requirement of rejection of images belonging clearly to neither of the trained classes (i.e. classification in a so-called 0-class). These special requirements are part of frameworks defined by central banks as the European Central Bank and are met by current ATMs and high speed sorting machines. We also consider training and classification time on state of the art GPU hardware. The study concentrates on the banknote recognition whereas banknote class dependent authenticity and fitness checks are a topic of its own which is not considered in this work.
With the abundance of data in recent years, interesting challenges are posed in the area of recommender systems. Producing high quality recommendations with scalability and performance is the need of the hour. Singular Value Decomposition(SVD) based recommendation algorithms have been leveraged to produce better results. In this paper, we extend the SVD technique further for scalability and performance in the context of 1) multi-threading 2) multiple computational units (with the use of Graphical Processing Units) and 3) distributed computation. We propose block based matrix factorization (BMF) paired with SVD. This enabled us to take advantage of SVD over basic matrix factorization(MF) while taking advantage of parallelism and scalability through BMF. We used Compute Unified Device Architecture (CUDA) platform and related hardware for leveraging Graphical Processing Unit (GPU) along with block based SVD to demonstrate the advantages in terms of performance and memory.
Breast lesion detection in ultrasound video is critical for computer-aided diagnosis. However, detecting lesion in video is quite challenging due to the blurred lesion boundary, high similarity to soft tissue and lack of video annotations. In this paper, we propose a semi-supervised breast lesion detection method based on temporal coherence which can detect the lesion more accurately. We aggregate features extracted from the historical key frames with adaptive key-frame scheduling strategy. Our proposed method accomplishes the unlabeled videos detection task by leveraging the supervision information from a different set of labeled images. In addition, a new WarpNet is designed to replace both the traditional spatial warping and feature aggregation operation, leading to a tremendous increase in speed. Experiments on 1,060 2D ultrasound sequences demonstrate that our proposed method achieves state-of-the-art video detection result as 91.3% in mean average precision and 19 ms per frame on GPU, compared to a RetinaNet based detection method in 86.6% and 32 ms.
Semantic Segmentation is an important module for autonomous robots such as self-driving cars. The advantage of video segmentation approaches compared to single image segmentation is that temporal image information is considered, and their performance increases due to this. Hence, single image segmentation approaches are extended by recurrent units such as convolutional LSTM (convLSTM) cells, which are placed at suitable positions in the basic network architecture. However, a major critique of video segmentation approaches based on recurrent neural networks is their large parameter count and their computational complexity, and so, their inference time of one video frame takes up to 66 percent longer than their basic version. Inspired by the success of the spatial and depthwise separable convolutional neural networks, we generalize these techniques for convLSTMs in this work, so that the number of parameters and the required FLOPs are reduced significantly. Experiments on different datasets show that the segmentation approaches using the proposed, modified convLSTM cells achieve similar or slightly worse accuracy, but are up to 15 percent faster on a GPU than the ones using the standard convLSTM cells. Furthermore, a new evaluation metric is introduced, which measures the amount of flickering pixels in the segmented video sequence.
The use of RGB-D information for salient object detection has been extensively explored in recent years. However, relatively few efforts have been put towards modeling salient object detection in real-world human activity scenes with RGBD. In this work, we fill the gap by making the following contributions to RGB-D salient object detection. (1) We carefully collect a new SIP (salient person) dataset, which consists of ~1K high-resolution images that cover diverse real-world scenes from various viewpoints, poses, occlusions, illuminations, and backgrounds. (2) We conduct a large-scale (and, so far, the most comprehensive) benchmark comparing contemporary methods, which has long been missing in the field and can serve as a baseline for future research. We systematically summarize 32 popular models and evaluate 18 parts of 32 models on seven datasets containing a total of about 97K images. (3) We propose a simple general architecture, called Deep Depth-Depurator Network (D3Net). It consists of a depth depurator unit (DDU) and a three-stream feature learning module (FLM), which performs low-quality depth map filtering and cross-modal feature learning respectively. These components form a nested structure and are elaborately designed to be learned jointly. D3Net exceeds the performance of any prior contenders across all five metrics under consideration, thus serving as a strong model to advance research in this field. We also demonstrate that D3Net can be used to efficiently extract salient object masks from real scenes, enabling effective background changing application with a speed of 65fps on a single GPU. All the saliency maps, our new SIP dataset, the D3Net model, and the evaluation tools are publicly available at https://github.com/DengPingFan/D3NetBenchmark.
Modern Automatic Speech Recognition (ASR) systems rely on distributed deep learning to for quick training completion. To enable efficient distributed training, it is imperative that the training algorithms can converge with a large mini-batch size. In this work, we discovered that Asynchronous Decentralized Parallel Stochastic Gradient Descent (ADPSGD) can work with much larger batch size than commonly used Synchronous SGD (SSGD) algorithm. On commonly used public SWB-300 and SWB-2000 ASR datasets, ADPSGD can converge with a batch size 3X as large as the one used in SSGD, thus enable training at a much larger scale. Further, we proposed a Hierarchical-ADPSGD (H-ADPSGD) system in which learners on the same computing node construct a super learner via a fast allreduce implementation, and super learners deploy ADPSGD algorithm among themselves. On a 64 Nvidia V100 GPU cluster connected via a 100Gb/s Ethernet network, our system is able to train SWB-2000 to reach a 7.6% WER on the Hub5-2000 Switchboard (SWB) test-set and a 13.2% WER on the Call-home (CH) test-set in 5.2 hours. To the best of our knowledge, this is the fastest ASR training system that attains this level of model accuracy for SWB-2000 task to be ever reported in the literature.
Differentiable architecture search (DARTS) provided a fast solution in finding effective network architectures, but suffered from large memory and computing overheads in jointly training a super-network and searching for an optimal architecture. In this paper, we present a novel approach, namely, Partially-Connected DARTS, by sampling a small part of super-network to reduce the redundancy in exploring the network space, thereby performing a more efficient search without comprising the performance. In particular, we perform operation search in a subset of channels while bypassing the held out part in a shortcut. This strategy may suffer from an undesired inconsistency on selecting the edges of super-net caused by sampling different channels. We alleviate it using edge normalization, which adds a new set of edge-level parameters to reduce uncertainty in search. Thanks to the reduced memory cost, PC-DARTS can be trained with a larger batch size and, consequently, enjoys both faster speed and higher training stability. Experimental results demonstrate the effectiveness of the proposed method. Specifically, we achieve an error rate of 2.57% on CIFAR10 with merely 0.1 GPU-days for architecture search, and a state-of-the-art top-1 error rate of 24.2% on ImageNet (under the mobile setting) using 3.8 GPU-days for search. Our code has been made available at: https://github.com/yuhuixu1993/PC-DARTS.
Purpose: Radio-frequency ablations play an important role in the therapy of malignant liver lesions. The navigation of a needle to the lesion poses a challenge for both the trainees and intervening physicians. Methods: This publication presents a new GPU-based, accurate method for the simulation of radio-frequency ablations for lesions at the needle tip in general and for an existing visuo-haptic 4D VR simulator. The method is implemented real-time capable with Nvidia CUDA. Results: It performs better than a literature method concerning the theoretical characteristic of monotonic convergence of the bioheat PDE and a in vitro gold standard with significant improvements (p < 0.05) in terms of Pearson correlations. It shows no failure modes or theoretically inconsistent individual simulation results after the initial phase of 10 seconds. On the Nvidia 1080 Ti GPU it achieves a very high frame rendering performance of >480 Hz. Conclusion: Our method provides a more robust and safer real-time ablation planning and intraoperative guidance technique, especially avoiding the over-estimation of the ablated tissue death zone, which is risky for the patient in terms of tumor recurrence. Future in vitro measurements and optimization shall further improve the conservative estimate.
The dominant approaches for named entity recognition (NER) mostly adopt complex recurrent neural networks (RNN), e.g., long-short-term-memory (LSTM). However, RNNs are limited by their recurrent nature in terms of computational efficiency. In contrast, convolutional neural networks (CNN) can fully exploit the GPU parallelism with their feedforward architectures. However, little attention has been paid to performing NER with CNNs, mainly owing to their difficulties in capturing the long-term context information in a sequence. In this paper, we propose a simple but effective CNN-based network for NER, i.e., gated relation network (GRN), which is more capable than common CNNs in capturing long-term context. Specifically, in GRN we firstly employ CNNs to explore the local context features of each word. Then we model the relations between words and use them as gates to fuse local context features into global ones for predicting labels. Without using recurrent layers that process a sentence in a sequential manner, our GRN allows computations to be performed in parallel across the entire sentence. Experiments on two benchmark NER datasets (i.e., CoNLL2003 and Ontonotes 5.0) show that, our proposed GRN can achieve state-of-the-art performance with or without external knowledge. It also enjoys lower time costs to train and test.We have made the code publicly available at https://github.com/HuiChen24/NER-GRN.
We present BlazeFace, a lightweight and well-performing face detector tailored for mobile GPU inference. It runs at a speed of 200-1000+ FPS on flagship devices. This super-realtime performance enables it to be applied to any augmented reality pipeline that requires an accurate facial region of interest as an input for task-specific models, such as 2D/3D facial keypoint or geometry estimation, facial features or expression classification, and face region segmentation. Our contributions include a lightweight feature extraction network inspired by, but distinct from MobileNetV1/V2, a GPU-friendly anchor scheme modified from Single Shot MultiBox Detector (SSD), and an improved tie resolution strategy alternative to non-maximum suppression.
GPUs are widely used to accelerate deep learning with NNs (NNs). On the other hand, since GPU memory capacity is limited, it is difficult to implement efficient programs that compute large NNs on GPU. To compute NNs exceeding GPU memory capacity, data-swapping method and recomputing method have been proposed in existing work. However, in these methods, performance overhead occurs due to data movement or increase of computation. In order to reduce the overhead, it is important to consider characteristics of each layer such as sizes and cost for recomputation. Based on this direction, we proposed Profiling based out-of-core Hybrid method (PoocH). PoocH determines target layers of swapping or recomputing based on runtime profiling. We implemented PoocH by extending a deep learning framework, Chainer, and we evaluated its performance. With PoocH, we successfully computed an NN requiring 50 GB memory on a single GPU with 16 GB memory. Compared with in-core cases, performance degradation was 38 \% on x86 machine and 28 \% on POWER9 machine.
In this paper, we propose Efficient Progressive Neural Architecture Search (EPNAS), a neural architecture search (NAS) that efficiently handles large search space through a novel progressive search policy with performance prediction based on REINFORCE~\cite{Williams.1992.PG}. EPNAS is designed to search target networks in parallel, which is more scalable on parallel systems such as GPU/TPU clusters. More importantly, EPNAS can be generalized to architecture search with multiple resource constraints, \eg, model size, compute complexity or intensity, which is crucial for deployment in widespread platforms such as mobile and cloud. We compare EPNAS against other state-of-the-art (SoTA) network architectures (\eg, MobileNetV2~\cite{mobilenetv2}) and efficient NAS algorithms (\eg, ENAS~\cite{pham2018efficient}, and PNAS~\cite{Liu2017b}) on image recognition tasks using CIFAR10 and ImageNet. On both datasets, EPNAS is superior \wrt architecture searching speed and recognition accuracy.
Depth from Small Motion (DfSM) (Ha et al., 2016) is particularly interesting for commercial handheld devices because it allows the possibility to get depth information with minimal user effort and cooperation. Due to speed and memory issue on these devices, the self calibration optimization of the method using Bundle Adjustment (BA) need as little as 10-15 images. Therefore, the optimization tends to take many iterations to converge or may not converge at all in some cases. This work propose a robust initialization for the bundle adjustment using the rank-1 factorization method (Tomasi and Kanade, 1992), (Aguiar and Moura, 1999a). We create a constraint matrix that is rank-1 in a noiseless situation, then use SVD to compute the inverse depth values and the camera motion. We only need about quarter fraction of the bundle adjustment iteration to converge. We also propose grided feature extraction technique so that only important and small features are tracked all over the image frames. This also ensure speedup in the full execution time on the mobile device. For the experiments, we have documented the execution time with the proposed Rank-1 initialization on two mobile device platforms using optimized accelerations with CPU-GPU co-processing. The combination of Rank 1-BA generates more robust depth-map and is significantly faster than using BA alone.
In this paper, we demonstrate a novel algorithm that uses ellipse fitting to estimate the bounding box rotation angle and size with the segmentation(mask) on the target for online and real-time visual object tracking. Our method, SiamMask_E, improves the bounding box fitting procedure of the state-of-the-art object tracking algorithm SiamMask and still retains a fast-tracking frame rate (80 fps) on a system equipped with GPU (GeForce GTX 1080 Ti or higher). We tested our approach on the visual object tracking datasets (VOT2016, VOT2018, and VOT2019) that were labeled with rotated bounding boxes. By comparing with the original SiamMask, we achieved an improved Accuracy of 0.652 and 0.309 EAO on VOT2019, which is 0.056 and 0.026 higher than the original SiamMask. The implementation is available on GitHub: https://github.com/baoxinchen/siammask_e.
Due to their prevalence, time series forecasting is crucial in multiple domains. We seek to make state-of-the-art forecasting fast, accessible, and generalizable. ES-RNN is a hybrid between classical state space forecasting models and modern RNNs that achieved a 9.4% sMAPE improvement in the M4 competition. Crucially, ES-RNN implementation requires per-time series parameters. By vectorizing the original implementation and porting the algorithm to a GPU, we achieve up to 322x training speedup depending on batch size with similar results as those reported in the original submission. Our code can be found at: https://github.com/damitkwr/ESRNN-GPU
We present Point-Voxel CNN (PVCNN) for efficient, fast 3D deep learning. Previous work processes 3D data using either voxel-based or point-based NN models. However, both approaches are computationally inefficient. The computation cost and memory footprints of the voxel-based models grow cubically with the input resolution, making it memory-prohibitive to scale up the resolution. As for point-based networks, up to 80% of the time is wasted on structuring the sparse data which have rather poor memory locality, not on the actual feature extraction. In this paper, we propose PVCNN that represents the 3D input data in points to reduce the memory consumption, while performing the convolutions in voxels to reduce the irregular, sparse data access and improve the locality. Our PVCNN model is both memory and computation efficient. Evaluated on semantic and part segmentation datasets, it achieves much higher accuracy than the voxel-based baseline with 10x GPU memory reduction; it also outperforms the state-of-the-art point-based models with 7x measured speedup on average. Remarkably, the narrower version of PVCNN achieves 2x speedup over PointNet (an extremely efficient model) on part and scene segmentation benchmarks with much higher accuracy. We validate the general effectiveness of PVCNN on 3D object detection: by replacing the primitives in Frustrum PointNet with PVConv, it outperforms Frustrum PointNet++ by 2.4% mAP on average with 1.5x measured speedup and GPU memory reduction.
We formulate object segmentation in video as a graph partitioning problem in space and time, in which nodes are pixels and their relations form local neighborhoods. We claim that the strongest cluster in this pixel-level graph represents the salient object segmentation. We compute the main cluster using a novel and fast 3D filtering technique that finds the spectral clustering solution, namely the principal eigenvector of the graph's adjacency matrix, without building the matrix explicitly - which would be intractable. Our method is based on the power iteration for finding the principal eigenvector of a matrix, which we prove is equivalent to performing a specific set of 3D convolutions in the space-time feature volume. This allows us to avoid creating the matrix and have a fast parallel implementation on GPU. We show that our method is much faster than classical power iteration applied directly on the adjacency matrix. Different from other works, ours is dedicated to preserving object consistency in space and time at the level of pixels. For that, it requires powerful pixel-wise features at the frame level. This makes it perfectly suitable for incorporating the output of a backbone network or other methods and fast-improving over their solution without supervision. In experiments, we obtain consistent improvement, with the same set of hyper-parameters, over the top state of the art methods on DAVIS-2016 dataset, both in unsupervised and semi-supervised tasks. We also achieve top results on the well-known SegTrackv2 dataset.
In this paper we tackle the problem of recovering the phase of complex linear measurements when only magnitude information is available and we control the input. We are motivated by the recent development of dedicated optics-based hardware for rapid random projections which leverages the propagation of light in random media. A signal of interest $\mathbf{\xi} \in \mathbb{R}^N$ is mixed by a random scattering medium to compute the projection $\mathbf{y} = \mathbf{A} \mathbf{\xi}$, with $\mathbf{A} \in \mathbb{C}^{M \times N}$ being a realization of a standard complex Gaussian iid random matrix. Such optics-based matrix multiplications can be much faster and energy-efficient than their CPU or GPU counterparts, yet two difficulties must be resolved: only the intensity ${|\mathbf{y}|}^2$ can be recorded by the camera, and the transmission matrix $\mathbf{A}$ is unknown. We show that even without knowing $\mathbf{A}$, we can recover the unknown phase of $\mathbf{y}$ for some equivalent transmission matrix with the same distribution as $\mathbf{A}$. Our method is based on two observations: first, conjugating or changing the phase of any row of $\mathbf{A}$ does not change its distribution; and second, since we control the input we can interfere $\mathbf{\xi}$ with arbitrary reference signals. We show how to leverage these observations to cast the measurement phase retrieval problem as a Euclidean distance geometry problem. We demonstrate appealing properties of the proposed algorithm in both numerical simulations and real hardware experiments. Not only does our algorithm accurately recover the missing phase, but it mitigates the effects of quantization and the sensitivity threshold, thus improving the measured magnitudes.
On-device inference of machine learning models for mobile phones is desirable due to its lower latency and increased privacy. Running such a compute-intensive task solely on the mobile CPU, however, can be difficult due to limited computing power, thermal constraints, and energy consumption. App developers and researchers have begun exploiting hardware accelerators to overcome these challenges. Recently, device manufacturers are adding neural processing units into high-end phones for on-device inference, but these account for only a small fraction of hand-held devices. In this paper, we present how we leverage the mobile GPU, a ubiquitous hardware accelerator on virtually every phone, to run inference of deep neural networks in real-time for both Android and iOS devices. By describing our architecture, we also discuss how to design networks that are mobile GPU-friendly. Our state-of-the-art mobile GPU inference engine is integrated into the open-source project TensorFlow Lite and publicly available at https://tensorflow.org/lite.