嵌入式和移动深度学习研究

冷不防 2024-04-17 19:33 41阅读 0赞

转载自：[https://github.com/EMDL/awesome-emdl][https_github.com_EMDL_awesome-emdl]

### 文章目录 ###

*  EMDL
 *   *  Papers
     *   *  Survey
         *  Model
         *  System
         *  Quantization
         *  Pruning
         *  Approximation
         *  Characterization
     *  Libraries
     *   *  Inference Framework
         *  Optimization Tools
         *  Research Demos
         *  Web
     *  Tutorials
     *   *  General
         *  NEON
         *  OpenCL
     *  Courses
     *  Demos
     *  General
     *   *  Vulkan
         *  OpenCL
         *  RenderScript
     *  Tools
     *   *  GPU
         *  Driver

# EMDL #

Embedded and mobile deep learning research notes

## Papers ##

### Survey ###

1.  [EfficientDNNs][]
2.  [A Survey of Model Compression and Acceleration for Deep Neural Networks][] \[arXiv '17\]

### Model ###

1.  [Searching for MobileNetV3][]\[arXiv '19, Google\]
2.  [MobilenetV2: Inverted Residuals and Linear Bottlenecks: Mobile Networks for  
    Classification, Detection and Segmentation][MobilenetV2_ Inverted Residuals and Linear Bottlenecks_ Mobile Networks for Classification_ Detection and Segmentation] \[arXiv '18, Google\]
3.  [NasNet: Learning Transferable Architectures for Scalable Image Recognition][NasNet_ Learning Transferable Architectures for Scalable Image Recognition] \[arXiv '17, Google\]
4.  [DeepRebirth: Accelerating Deep Neural Network Execution on Mobile Devices][DeepRebirth_ Accelerating Deep Neural Network Execution on Mobile Devices] \[AAAI’18, Samsung\]
5.  [ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices][ShuffleNet_ An Extremely Efficient Convolutional Neural Network for Mobile Devices] \[arXiv '17, Megvii\]
6.  [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications][MobileNets_ Efficient Convolutional Neural Networks for Mobile Vision Applications] \[arXiv '17, Google\]
7.  [CondenseNet: An Efficient DenseNet using Learned Group Convolutions][CondenseNet_ An Efficient DenseNet using Learned Group Convolutions] \[arXiv '17\]

### System ###

1.  [DeepMon: Mobile GPU-based Deep Learning Framework for Continuous Vision Applications][DeepMon_ Mobile GPU-based Deep Learning Framework for Continuous Vision Applications] \[MobiSys '17\]
2.  [DeepEye: Resource Efficient Local Execution of Multiple Deep Vision Models using Wearable Commodity Hardware][DeepEye_ Resource Efficient Local Execution of Multiple Deep Vision Models using Wearable Commodity Hardware] \[MobiSys '17\]
3.  [MobiRNN: Efficient Recurrent Neural Network Execution on Mobile GPU][MobiRNN_ Efficient Recurrent Neural Network Execution on Mobile GPU] \[EMDL '17\]
4.  [DeepSense: A GPU-based deep convolutional neural network framework on commodity mobile devices][DeepSense_ A GPU-based deep convolutional neural network framework on commodity mobile devices] \[WearSys '16\]
5.  [DeepX: A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices][DeepX_ A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices] \[IPSN '16\]
6.  [EIE: Efficient Inference Engine on Compressed Deep Neural Network][EIE_ Efficient Inference Engine on Compressed Deep Neural Network] \[ISCA '16\]
7.  [MCDNN: An Approximation-Based Execution Framework for Deep Stream Processing Under Resource Constraints][MCDNN_ An Approximation-Based Execution Framework for Deep Stream Processing Under Resource Constraints] \[MobiSys '16\]
8.  [DXTK: Enabling Resource-efficient Deep Learning on Mobile and Embedded Devices with the DeepX Toolkit][DXTK_ Enabling Resource-efficient Deep Learning on Mobile and Embedded Devices with the DeepX Toolkit] \[MobiCASE '16\]
9.  [Sparsification and Separation of Deep Learning Layers for Constrained Resource Inference on Wearables][] \[SenSys ’16\]
10. [An Early Resource Characterization of Deep Learning on Wearables, Smartphones and Internet-of-Things Devices][An Early Resource Characterization of Deep Learning on Wearables_ Smartphones and Internet-of-Things Devices] \[IoT-App ’15\]
11. [CNNdroid: GPU-Accelerated Execution of Trained Deep Convolutional Neural Networks on Android][CNNdroid_ GPU-Accelerated Execution of Trained Deep Convolutional Neural Networks on Android] \[MM '16\]
12. [fpgaConvNet: A Toolflow for Mapping Diverse Convolutional Neural Networks on Embedded FPGAs][fpgaConvNet_ A Toolflow for Mapping Diverse Convolutional Neural Networks on Embedded FPGAs] \[NIPS '17\]

### Quantization ###

1.  [Quantizing deep convolutional networks for efficient inference: A whitepaper][Quantizing deep convolutional networks for efficient inference_ A whitepaper]
2.  [LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks][LQ-Nets_ Learned Quantization for Highly Accurate and Compact Deep Neural Networks] \[ECCV’18\]
3.  [The ZipML Framework for Training Models with End-to-End Low Precision: The Cans, the Cannots, and a Little Bit of Deep Learning][The ZipML Framework for Training Models with End-to-End Low Precision_ The Cans_ the Cannots_ and a Little Bit of Deep Learning] \[ICML’17\]
4.  [Compressing Deep Convolutional Networks using Vector Quantization][] \[arXiv’14\]
5.  [Quantized Convolutional Neural Networks for Mobile Devices][] \[CVPR '16\]
6.  [Fixed-Point Performance Analysis of Recurrent Neural Networks][] \[ICASSP’16\]
7.  [Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations][Quantized Neural Networks_ Training Neural Networks with Low Precision Weights and Activations] \[arXiv’16\]
8.  [Loss-aware Binarization of Deep Networks][] \[ICLR’17\]
9.  [Towards the Limit of Network Quantization][] \[ICLR’17\]
10. [Deep Learning with Low Precision by Half-wave Gaussian Quantization][] \[CVPR’17\]
11. [ShiftCNN: Generalized Low-Precision Architecture for Inference of Convolutional Neural Networks][ShiftCNN_ Generalized Low-Precision Architecture for Inference of Convolutional Neural Networks] \[arXiv’17\]
12. [Training and Inference with Integers in Deep Neural Networks][] \[ICLR’18\]

### Pruning ###

1.  [Awesome-Pruning][]
2.  [Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration][] \[CVPR’19\]
3.  [Learning both Weights and Connections for Efficient Neural Networks][] \[NIPS’15\]
4.  [Pruning Filters for Efficient ConvNets][] \[ICLR’17\]
5.  [Pruning Convolutional Neural Networks for Resource Efficient Inference][] \[ICLR’17\]
6.  [Soft Weight-Sharing for Neural Network Compression][] \[ICLR’17\]
7.  [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding][Deep Compression_ Compressing Deep Neural Networks with Pruning_ Trained Quantization and Huffman Coding] \[ICLR’16\]
8.  [Dynamic Network Surgery for Efficient DNNs][] \[NIPS’16\]
9.  [Designing Energy-Efficient Convolutional Neural Networks using Energy-Aware Pruning][] \[CVPR’17\]
10. [ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression][ThiNet_ A Filter Level Pruning Method for Deep Neural Network Compression] \[ICCV’17\]
11. [To prune, or not to prune: exploring the efficacy of pruning for model compression][To prune_ or not to prune_ exploring the efficacy of pruning for model compression] \[ICLR’18\]

### Approximation ###

1.  [Efficient and Accurate Approximations of Nonlinear Convolutional Networks][] \[CVPR’15\]
2.  [Accelerating Very Deep Convolutional Networks for Classification and Detection][] (Extended version of above one)
3.  [Convolutional neural networks with low-rank regularization][] \[arXiv’15\]
4.  [Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation][] \[NIPS’14\]
5.  [Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications][] \[ICLR’16\]
6.  [High performance ultra-low-precision convolutions on mobile devices][] \[NIPS’17\]

### Characterization ###

1.  [A First Look at Deep Learning Apps on Smartphones][] \[WWW’19\]
2.  [Machine Learning at Facebook: Understanding Inference at the Edge][Machine Learning at Facebook_ Understanding Inference at the Edge] \[HPCA’19\]
3.  [NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications][NetAdapt_ Platform-Aware Neural Network Adaptation for Mobile Applications] \[ECCV 2018\]
4.  [Latency and Throughput Characterization of Convolutional Neural Networks for Mobile Computer Vision][] \[MMSys’18\]

## Libraries ##

### Inference Framework ###

1.  [alibaba/MNN][alibaba_MNN]
2.  [TensorFlow Lite GPU][]
3.  [TensorFlow Lite][]
4.  [XiaoMi/mace: MACE is a deep learning inference framework optimized for mobile heterogeneous computing platforms.][XiaoMi_mace_ MACE is a deep learning inference framework optimized for mobile heterogeneous computing platforms.]
5.  [Tencent/ncnn: ncnn is a high-performance neural network inference framework optimized for the mobile platform][Tencent_ncnn_ ncnn is a high-performance neural network inference framework optimized for the mobile platform]
6.  [baidu/paddle-mobile][baidu_paddle-mobile]
7.  [Apple CoreML][]
8.  [Snapdragon Neural Processing Engine][]
9.  [ARM-software/ComputeLibrary: The ARM Computer Vision and Machine Learning library is a set of functions optimised for both ARM CPUs and GPUs using SIMD technologies][ARM-software_ComputeLibrary_ The ARM Computer Vision and Machine Learning library is a set of functions optimised for both ARM CPUs and GPUs using SIMD technologies], [Intro][]
10. [Microsoft Embedded Learning Library][]
11. [MXNet Amalgamation][]
12. [OAID/Tengine: Tengine is a lite, high performance, modular inference engine for embedded device][OAID_Tengine_ Tengine is a lite_ high performance_ modular inference engine for embedded device]
13. [xmartlabs/Bender: Easily craft fast Neural Networks on iOS! Use TensorFlow models. Metal under the hood.][xmartlabs_Bender_ Easily craft fast Neural Networks on iOS_ Use TensorFlow models. Metal under the hood.]
14. [JDAI-CV/dabnn: dabnn is an accelerated binary neural networks inference framework for mobile platform][JDAI-CV_dabnn_ dabnn is an accelerated binary neural networks inference framework for mobile platform]

### Optimization Tools ###

1.  [Neural Network Distiller][]
2.  [An Automatic Model Compression (AutoMC) framework for developing smaller and faster AI applications][An Automatic Model Compression _AutoMC_ framework for developing smaller and faster AI applications]

### Research Demos ###

1.  [RSTensorFlow: GPU Accelerated TensorFlow for Commodity Android Devices][RSTensorFlow_ GPU Accelerated TensorFlow for Commodity Android Devices]

### Web ###

1.  [mil-tokyo/webdnn: Fastest DNN Execution Framework on Web Browser][mil-tokyo_webdnn_ Fastest DNN Execution Framework on Web Browser]

## Tutorials ##

### General ###

1.  [Squeezing Deep Learning Into Mobile Phones][]
2.  [Deep Learning – Tutorial and Recent Trends][Deep Learning _ Tutorial and Recent Trends]
3.  [Tutorial on Hardware Architectures for Deep Neural Networks][]
4.  [Efficient Convolutional Neural Network Inference on Mobile GPUs][]

### NEON ###

1.  [NEON™ Programmer’s Guide][NEON_ Programmer_s Guide]

### OpenCL ###

1.  [ARM® Mali™ GPU OpenCL Developer Guide][ARM_ Mali_ GPU OpenCL Developer Guide], [pdf][]
2.  [Optimal Compute on ARM Mali™ GPUs][Optimal Compute on ARM Mali_ GPUs]
3.  [GPU Compute for Mobile Devices][]
4.  [Compute for Mobile Devices Performance focused][]
5.  [Hands On OpenCL][]
6.  [Adreno OpenCL Programming Guide][]
7.  [Better OpenCL Performance on Qualcomm Adreno GPU][]

## Courses ##

1.  [Deep learning **systems**][Deep learning _systems], UW course schedule(focused on systems design, not learning)

## Demos ##

## General ##

1.  [TensorFlow Android Camera Demo][]
2.  [TensorFlow iOS Example][]
3.  [Caffe2 AICamera][]

### Vulkan ###

1.  [Vulkan API Examples and Demos][]
2.  [Neural Machine Translation on Android][]

### OpenCL ###

1.  [DeepMon][]

### RenderScript ###

1.  [Mobile\_ConvNet: RenderScript CNN for Android][Mobile_ConvNet_ RenderScript CNN for Android]

## Tools ##

### GPU ###

1.  [Bifrost GPU architecture and ARM Mali-G71 GPU][]
2.  [Midgard GPU Architecture][], [ARM Mali-T880 GPU][]
3.  [Mobile GPU market share][]

### Driver ###

1.  \[Adreno\] [csarron/qcom\_vendor\_binaries: Common Proprietary Qualcomm Binaries][csarron_qcom_vendor_binaries_ Common Proprietary Qualcomm Binaries]
2.  \[Mali\] [Fevax/vendor\_samsung\_hero2ltexx: Blobs from s7 Edge G935F][Fevax_vendor_samsung_hero2ltexx_ Blobs from s7 Edge G935F]

[https_github.com_EMDL_awesome-emdl]: https://github.com/EMDL/awesome-emdl
[EfficientDNNs]: https://github.com/MingSun-Tse/EfficientDNNs
[A Survey of Model Compression and Acceleration for Deep Neural Networks]: https://arxiv.org/abs/1710.09282
[Searching for MobileNetV3]: https://arxiv.org/abs/1905.02244
[MobilenetV2_ Inverted Residuals and Linear Bottlenecks_ Mobile Networks for Classification_ Detection and Segmentation]: https://arxiv.org/pdf/1801.04381.pdf
[NasNet_ Learning Transferable Architectures for Scalable Image Recognition]: https://arxiv.org/pdf/1707.07012.pdf
[DeepRebirth_ Accelerating Deep Neural Network Execution on Mobile Devices]: https://arxiv.org/abs/1708.04728
[ShuffleNet_ An Extremely Efficient Convolutional Neural Network for Mobile Devices]: https://arxiv.org/abs/1707.01083
[MobileNets_ Efficient Convolutional Neural Networks for Mobile Vision Applications]: https://arxiv.org/abs/1704.04861
[CondenseNet_ An Efficient DenseNet using Learned Group Convolutions]: https://arxiv.org/abs/1711.09224
[DeepMon_ Mobile GPU-based Deep Learning Framework for Continuous Vision Applications]: https://www.sigmobile.org/mobisys/2017/accepted.php
[DeepEye_ Resource Efficient Local Execution of Multiple Deep Vision Models using Wearable Commodity Hardware]: http://fahim-kawsar.net/papers/Mathur.MobiSys2017-Camera.pdf
[MobiRNN_ Efficient Recurrent Neural Network Execution on Mobile GPU]: https://arxiv.org/abs/1706.00878
[DeepSense_ A GPU-based deep convolutional neural network framework on commodity mobile devices]: http://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=4278&context=sis_research
[DeepX_ A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices]: http://niclane.org/pubs/deepx_ipsn.pdf
[EIE_ Efficient Inference Engine on Compressed Deep Neural Network]: https://arxiv.org/abs/1602.01528
[MCDNN_ An Approximation-Based Execution Framework for Deep Stream Processing Under Resource Constraints]: http://haneul.github.io/papers/mcdnn.pdf
[DXTK_ Enabling Resource-efficient Deep Learning on Mobile and Embedded Devices with the DeepX Toolkit]: http://niclane.org/pubs/dxtk_mobicase.pdf
[Sparsification and Separation of Deep Learning Layers for Constrained Resource Inference on Wearables]: http://niclane.org/pubs/sparsesep_sensys.pdf
[An Early Resource Characterization of Deep Learning on Wearables_ Smartphones and Internet-of-Things Devices]: http://niclane.org/pubs/iotapp15_early.pdf
[CNNdroid_ GPU-Accelerated Execution of Trained Deep Convolutional Neural Networks on Android]: https://arxiv.org/abs/1511.07376
[fpgaConvNet_ A Toolflow for Mapping Diverse Convolutional Neural Networks on Embedded FPGAs]: https://arxiv.org/abs/1711.08740
[Quantizing deep convolutional networks for efficient inference_ A whitepaper]: https://arxiv.org/abs/1806.08342
[LQ-Nets_ Learned Quantization for Highly Accurate and Compact Deep Neural Networks]: https://arxiv.org/pdf/1807.10029.pdf
[The ZipML Framework for Training Models with End-to-End Low Precision_ The Cans_ the Cannots_ and a Little Bit of Deep Learning]: https://arxiv.org/abs/1611.05402
[Compressing Deep Convolutional Networks using Vector Quantization]: https://arxiv.org/abs/1412.6115
[Quantized Convolutional Neural Networks for Mobile Devices]: https://arxiv.org/abs/1512.06473
[Fixed-Point Performance Analysis of Recurrent Neural Networks]: https://arxiv.org/abs/1512.01322
[Quantized Neural Networks_ Training Neural Networks with Low Precision Weights and Activations]: https://arxiv.org/abs/1609.07061
[Loss-aware Binarization of Deep Networks]: https://arxiv.org/abs/1611.01600
[Towards the Limit of Network Quantization]: https://arxiv.org/abs/1612.01543
[Deep Learning with Low Precision by Half-wave Gaussian Quantization]: https://arxiv.org/abs/1702.00953
[ShiftCNN_ Generalized Low-Precision Architecture for Inference of Convolutional Neural Networks]: https://arxiv.org/abs/1706.02393
[Training and Inference with Integers in Deep Neural Networks]: https://openreview.net/forum?id=HJGXzmspb
[Awesome-Pruning]: https://github.com/he-y/Awesome-Pruning
[Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration]: https://arxiv.org/abs/1811.00250
[Learning both Weights and Connections for Efficient Neural Networks]: https://arxiv.org/abs/1506.02626
[Pruning Filters for Efficient ConvNets]: https://arxiv.org/abs/1608.08710
[Pruning Convolutional Neural Networks for Resource Efficient Inference]: https://arxiv.org/abs/1611.06440
[Soft Weight-Sharing for Neural Network Compression]: https://arxiv.org/abs/1702.04008
[Deep Compression_ Compressing Deep Neural Networks with Pruning_ Trained Quantization and Huffman Coding]: https://arxiv.org/abs/1510.00149
[Dynamic Network Surgery for Efficient DNNs]: https://arxiv.org/abs/1608.04493
[Designing Energy-Efficient Convolutional Neural Networks using Energy-Aware Pruning]: https://arxiv.org/abs/1611.05128
[ThiNet_ A Filter Level Pruning Method for Deep Neural Network Compression]: https://arxiv.org/abs/1707.06342
[To prune_ or not to prune_ exploring the efficacy of pruning for model compression]: https://arxiv.org/abs/1710.01878
[Efficient and Accurate Approximations of Nonlinear Convolutional Networks]: https://arxiv.org/abs/1411.4229
[Accelerating Very Deep Convolutional Networks for Classification and Detection]: https://arxiv.org/abs/1505.06798
[Convolutional neural networks with low-rank regularization]: https://arxiv.org/abs/1511.06067
[Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation]: https://arxiv.org/abs/1404.0736
[Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications]: https://arxiv.org/abs/1511.06530
[High performance ultra-low-precision convolutions on mobile devices]: https://arxiv.org/abs/1712.02427
[A First Look at Deep Learning Apps on Smartphones]: https://arxiv.org/abs/1812.05448
[Machine Learning at Facebook_ Understanding Inference at the Edge]: https://research.fb.com/publications/machine-learning-at-facebook-understanding-inference-at-the-edge/
[NetAdapt_ Platform-Aware Neural Network Adaptation for Mobile Applications]: https://arxiv.org/abs/1804.03230
[Latency and Throughput Characterization of Convolutional Neural Networks for Mobile Computer Vision]: https://arxiv.org/abs/1803.09492
[alibaba_MNN]: https://github.com/alibaba/MNN
[TensorFlow Lite GPU]: https://www.tensorflow.org/lite/performance/gpu
[TensorFlow Lite]: https://www.tensorflow.org/mobile/tflite/
[XiaoMi_mace_ MACE is a deep learning inference framework optimized for mobile heterogeneous computing platforms.]: https://github.com/XiaoMi/mace
[Tencent_ncnn_ ncnn is a high-performance neural network inference framework optimized for the mobile platform]: https://github.com/Tencent/ncnn
[baidu_paddle-mobile]: https://github.com/PaddlePaddle/paddle-mobile
[Apple CoreML]: https://developer.apple.com/documentation/coreml
[Snapdragon Neural Processing Engine]: https://developer.qualcomm.com/software/snapdragon-neural-processing-engine
[ARM-software_ComputeLibrary_ The ARM Computer Vision and Machine Learning library is a set of functions optimised for both ARM CPUs and GPUs using SIMD technologies]: https://github.com/ARM-software/ComputeLibrary
[Intro]: https://developer.arm.com/technologies/compute-library
[Microsoft Embedded Learning Library]: https://github.com/Microsoft/ELL
[MXNet Amalgamation]: https://github.com/dmlc/mxnet/tree/master/amalgamation
[OAID_Tengine_ Tengine is a lite_ high performance_ modular inference engine for embedded device]: https://github.com/OAID/Tengine
[xmartlabs_Bender_ Easily craft fast Neural Networks on iOS_ Use TensorFlow models. Metal under the hood.]: https://github.com/xmartlabs/Bender
[JDAI-CV_dabnn_ dabnn is an accelerated binary neural networks inference framework for mobile platform]: https://github.com/JDAI-CV/dabnn
[Neural Network Distiller]: https://github.com/NervanaSystems/distiller
[An Automatic Model Compression _AutoMC_ framework for developing smaller and faster AI applications]: https://github.com/Tencent/PocketFlow
[RSTensorFlow_ GPU Accelerated TensorFlow for Commodity Android Devices]: https://nesl.github.io/RSTensorFlow/
[mil-tokyo_webdnn_ Fastest DNN Execution Framework on Web Browser]: https://github.com/mil-tokyo/webdnn
[Squeezing Deep Learning Into Mobile Phones]: https://www.slideshare.net/anirudhkoul/squeezing-deep-learning-into-mobile-phones
[Deep Learning _ Tutorial and Recent Trends]: https://www.dropbox.com/s/p7lvelt0aihrwtl/FPGA%2717%20tutorial%20Song%20Han.pdf?dl=0
[Tutorial on Hardware Architectures for Deep Neural Networks]: http://eyeriss.mit.edu/tutorial.html
[Efficient Convolutional Neural Network Inference on Mobile GPUs]: https://www.slideshare.net/embeddedvision/efficient-convolutional-neural-network-inference-on-mobile-gpus-a-presentation-from-imagination-technologies
[NEON_ Programmer_s Guide]: https://developer.arm.com/docs/den0018/latest/neontm-version-10-programmers-guide
[ARM_ Mali_ GPU OpenCL Developer Guide]: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.100614_0303_00_en/ada1432742770595.html
[pdf]: http://infocenter.arm.com/help/topic/com.arm.doc.100614_0303_00_en/arm_mali_gpu_opencl_developer_guide_100614_0303_00_en.pdf
[Optimal Compute on ARM Mali_ GPUs]: http://www.cs.bris.ac.uk/home/simonm/montblanc/OpenCL_on_Mali.pdf
[GPU Compute for Mobile Devices]: http://www.iwocl.org/wp-content/uploads/iwocl-2014-workshop-Tim-Hartley.pdf
[Compute for Mobile Devices Performance focused]: http://kesen.realtimerendering.com/Compute_for_Mobile_Devices5.pdf
[Hands On OpenCL]: https://handsonopencl.github.io/
[Adreno OpenCL Programming Guide]: https://developer.qualcomm.com/download/adrenosdk/adreno-opencl-programming-guide.pdf
[Better OpenCL Performance on Qualcomm Adreno GPU]: https://developer.qualcomm.com/blog/better-opencl-performance-qualcomm-adreno-gpu-memory-optimization
[Deep learning _systems]: http://dlsys.cs.washington.edu/schedule
[TensorFlow Android Camera Demo]: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/examples/android
[TensorFlow iOS Example]: https://github.com/hollance/TensorFlow-iOS-Example
[Caffe2 AICamera]: https://github.com/bwasti/AICamera
[Vulkan API Examples and Demos]: https://github.com/SaschaWillems/Vulkan
[Neural Machine Translation on Android]: https://github.com/harvardnlp/nmt-android
[DeepMon]: https://github.com/JC1DA/DeepMon
[Mobile_ConvNet_ RenderScript CNN for Android]: https://github.com/mtmd/Mobile_ConvNet
[Bifrost GPU architecture and ARM Mali-G71 GPU]: https://www.hotchips.org/wp-content/uploads/hc_archives/hc28/HC28.22-Monday-Epub/HC28.22.10-GPU-HPC-Epub/HC28.22.110-Bifrost-JemDavies-ARM-v04-9.pdf
[Midgard GPU Architecture]: http://malideveloper.arm.com/downloads/ARM_Game_Developer_Days/PDFs/2-Mali-GPU-architecture-overview-and-tile-local-storage.pdf
[ARM Mali-T880 GPU]: https://www.hotchips.org/wp-content/uploads/hc_archives/hc27/HC27.25-Tuesday-Epub/HC27.25.50-GPU-Epub/HC27.25.531-Mali-T880-Bratt-ARM-2015_08_23.pdf
[Mobile GPU market share]: https://hwstats.unity3d.com/mobile/gpu.html
[csarron_qcom_vendor_binaries_ Common Proprietary Qualcomm Binaries]: https://github.com/csarron/qcom_vendor_binaries
[Fevax_vendor_samsung_hero2ltexx_ Blobs from s7 Edge G935F]: https://github.com/Fevax/vendor_samsung_hero2ltexx