Merge pull request #46 from JDAI-CV/update_readme

Update README furtherly for our arxiv eprint

README.md

@@ -12,6 +12,8 @@
 
 Join chat at [Gitter (English)](https://gitter.im/dabnn/dabnn) or QQ Group (Chinese, 1021964010, answer: nndab)
 
+Our ACM MM paper: https://arxiv.org/abs/1908.05858
+
 ## Introduction
 
 Binary neural networks (BNNs) have great potential on edge devices since they replace float operations by efficient bit-wise operations. However, to leverage the efficiency of bit-wise operations, the reimplmentation of convolution layer and also other layers is needed. 
@@ -37,7 +39,7 @@ dabnn_3x3_256                    975123 ns     969810 ns        691     <--- inp
 dabnn_3x3_256_s2                 268310 ns     267712 ns       2618     <--- input: 14*14*256, kernel: 256*3*3*256, output: 7*7*256, padding: 1, stride: 2
 dabnn_3x3_512                   1281832 ns    1253921 ns        588     <--- input:  7* 7*512, kernel: 512*3*3*512, output:  7* 7*512, padding: 1
 dabnn_bireal18_imagenet        61920154 ns   61339185 ns         10     <--- Bi-Real Net 18, 56.4% top-1 on ImageNet
-dabnn_bireal18_imagenet_stem   43294019 ns   41401923 ns         14     <--- Bi-Real Net 18 with stem module (The network structure is described in detail in our paper https://arxiv.org/abs/1908.05858), 56.4% top-1 on ImageNet
+dabnn_bireal18_imagenet_stem   43294019 ns   41401923 ns         14     <--- Bi-Real Net 18 with stem module (The network structure is described in detail in [our paper](https://arxiv.org/abs/1908.05858)), 56.4% top-1 on ImageNet
 ```
 
 The following is the comparison between our dabnn and [Caffe](http://caffe.berkeleyvision.org) (full precision), [TensorFlow Lite](https://www.tensorflow.org/lite) (full precision) and [BMXNet](https://github.com/hpi-xnor/BMXNet) (binary). We surprisingly observe that BMXNet is even slower than the full precision TensorFlow Lite. It suggests that the potential of binary neural networks is far from exploited until our dabnn is published.
@@ -64,14 +66,16 @@ We publish two pretrained binary neural network models based on [Bi-Real Net](ht
 
 * Bi-Real Net 18, 56.4% top-1 on ImageNet, 61.3ms/image on Google Pixel 1 (single thread). [[dabnn](https://drive.google.com/uc?export=download&id=1Oau5CtFR9nWXmlBBU47Jg5ypMiIEMtvo)] [[ONNX](https://drive.google.com/uc?export=download&id=1Xp3HB51H6Nhl6e555ieJubVutQake5sR)]
 
-* Bi-Real Net 18 with Stem Module, 56.4% top-1 on ImageNet, 43.2ms/image on Google Pixel 1 (single thread). The detailed network structure is described our paper https://arxiv.org/abs/1908.05858. [[dabnn](https://drive.google.com/uc?export=download&id=1ArsirMdbtJ9lvHSjc1hkQ7dIXDKh-D1t)] [[ONNX](https://drive.google.com/uc?export=download&id=1zu48CFptAGZ91IDCBPJSPM0bxDuPm9HS)]
+* Bi-Real Net 18 with Stem Module, 56.4% top-1 on ImageNet, 43.2ms/image on Google Pixel 1 (single thread). The detailed network structure is described in [our paper](https://arxiv.org/abs/1908.05858). [[dabnn](https://drive.google.com/uc?export=download&id=1ArsirMdbtJ9lvHSjc1hkQ7dIXDKh-D1t)] [[ONNX](https://drive.google.com/uc?export=download&id=1zu48CFptAGZ91IDCBPJSPM0bxDuPm9HS)]
 
 ## Implementation Details
 
 * The Implementation of Binary Convolutions: [docs/bconv.md](docs/bconv.md)
 
 * Model Conversion: [docs/onnx2bnn.md](docs/onnx2bnn.md)
 
+For more details please read [our ACM MM paper](https://arxiv.org/abs/1908.05858).
+
 ## Example project
 
 Android app demo: https://github.com/JDAI-CV/dabnn-example

README_CN.md

@@ -12,6 +12,8 @@ Gitter (English): [dabnn/dabnn](https://gitter.im/dabnn/dabnn)
 
 QQ 群：1021964010, 入群答案: nndab
 
+dabnn 已经被多媒体领域的会议 ACM MM 接收: https://arxiv.org/abs/1908.05858
+
 ## 简介
 
 二值网络（Binary Neural Networks，又称 1-bit 网络）是一种特殊的神经网络，它将网络的权重和中间特征压缩为 1 个 bit，可以看作是普通浮点型网络量化到极致的结果。和其它 bit 数稍高的量化网络（例如三值网络、2-bit 网络、4-bit 网络）相比，二值网络最突出的优点在于，1-bit 乘加操作可以通过位运算高效实现，因此可以无缝运行在主流硬件平台（x86、ARM）上，而三值、2-bit、4-bit 等等操作需要特殊硬件平台（FPGA、ASIC）支持。在计算效率上，一个 32 位浮点运算指令只能完成一次浮点乘法或加法运算，但一个 32 位位运算指令可以同时完成 32 个 1-bit 乘法运算。
@@ -39,7 +41,7 @@ dabnn_3x3_256                    975123 ns     969810 ns        691     <--- inp
 dabnn_3x3_256_s2                 268310 ns     267712 ns       2618     <--- input: 14*14*256, kernel: 256*3*3*256, output: 7*7*256, padding: 1, stride: 2
 dabnn_3x3_512                   1281832 ns    1253921 ns        588     <--- input:  7* 7*512, kernel: 512*3*3*512, output:  7* 7*512, padding: 1
 dabnn_bireal18_imagenet        61920154 ns   61339185 ns         10     <--- Bi-Real Net 18, ImageNet top-1 为 56.4%
-dabnn_bireal18_imagenet_stem   43294019 ns   41401923 ns         14     <--- 带有 stem 模块的 Bi-Real Net 18 (在我们的 paper https://arxiv.org/abs/1908.05858 中有描述), ImageNet top-1 为 56.4%
+dabnn_bireal18_imagenet_stem   43294019 ns   41401923 ns         14     <--- 带有 stem 模块的 Bi-Real Net 18 (在 [我们的 paper](https://arxiv.org/abs/1908.05858) 中有描述), ImageNet top-1 为 56.4%
 ```
 
 在 Google Pixel 1 上与 [Caffe](http://caffe.berkeleyvision.org)（全精度）, [TensorFlow Lite](https://www.tensorflow.org/lite)（全精度）和 [BMXNet](https://github.com/hpi-xnor/BMXNet)（二值）的对比如下。我们很惊讶的发现现有的二值 inference 框架 BMXNet 甚至比全精度的 TensorFlow Lite 还要慢，这表明，直到 dabnn 推出之前，二值网络的潜力都远远没有被挖掘出来。
@@ -66,14 +68,16 @@ dabnn_bireal18_imagenet_stem   43294019 ns   41401923 ns         14     <--- 带
 
 * Bi-Real Net 18, ImageNet top-1 为 56.4%, 在 Google Pixel 1 上的速度为 61.8ms/image （单线程）。[[dabnn](https://drive.google.com/uc?export=download&id=1Oau5CtFR9nWXmlBBU47Jg5ypMiIEMtvo)] [[ONNX](https://drive.google.com/uc?export=download&id=1Xp3HB51H6Nhl6e555ieJubVutQake5sR)]
 
-* Bi-Real Net 18 with Stem Module, ImageNet top-1 为 56.4%, 在 Google Pixel 1 上的速度为 43.2ms/image （单线程）。详细的网络结构在我们的 paper https://arxiv.org/abs/1908.05858 中有描述。 [[dabnn](https://drive.google.com/uc?export=download&id=1ArsirMdbtJ9lvHSjc1hkQ7dIXDKh-D1t)] [[ONNX](https://drive.google.com/uc?export=download&id=1zu48CFptAGZ91IDCBPJSPM0bxDuPm9HS)]
+* Bi-Real Net 18 with Stem Module, ImageNet top-1 为 56.4%, 在 Google Pixel 1 上的速度为 43.2ms/image （单线程）。详细的网络结构在[我们的 paper](https://arxiv.org/abs/1908.05858) 中有描述。 [[dabnn](https://drive.google.com/uc?export=download&id=1ArsirMdbtJ9lvHSjc1hkQ7dIXDKh-D1t)] [[ONNX](https://drive.google.com/uc?export=download&id=1zu48CFptAGZ91IDCBPJSPM0bxDuPm9HS)]
 
 ## 技术细节
 
 * Binary Convolutions 的实现: [docs/bconv.md](docs/bconv_CN.md)
 
 * 模型转换: [docs/onnx2bnn.md](docs/onnx2bnn_CN.md)
 
+更多细节可以参阅[我们的 ACM MM paper](https://arxiv.org/abs/1908.05858)
+
 ## 示例工程
 
 Android app demo: https://github.com/JDAI-CV/dabnn-example