From patchwork Thu Jun 18 09:15:35 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Fu, Ting" X-Patchwork-Id: 20457 Return-Path: X-Original-To: patchwork@ffaux-bg.ffmpeg.org Delivered-To: patchwork@ffaux-bg.ffmpeg.org Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by ffaux.localdomain (Postfix) with ESMTP id D23E744B58B for ; Thu, 18 Jun 2020 12:21:53 +0300 (EEST) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id C08FE68B780; Thu, 18 Jun 2020 12:21:53 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mga07.intel.com (mga07.intel.com [134.134.136.100]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 4167B68B74D for ; Thu, 18 Jun 2020 12:21:47 +0300 (EEST) IronPort-SDR: iNT83czEQFpSAXER7ySepI+sYHjWLHWNZ7H3Bfm+RALH8YQK0R1BLdB7aC5GrfyAQW5SY3J0Di SiuEpK+MeX3A== X-IronPort-AV: E=McAfee;i="6000,8403,9655"; a="207748157" X-IronPort-AV: E=Sophos;i="5.73,526,1583222400"; d="scan'208";a="207748157" X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga006.fm.intel.com ([10.253.24.20]) by orsmga105.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 18 Jun 2020 02:21:27 -0700 IronPort-SDR: BqlUgsQsXwyct0mzBDr8mwB7i7bD81KUHbOO+bwLKv1R1VO9qONgwOvJP3J9cVtm0Mew2dFpjY Ui/2lYJ9QH/g== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.73,526,1583222400"; d="scan'208";a="477130708" Received: from semmer-ubuntu.sh.intel.com ([10.239.159.126]) by fmsmga006.fm.intel.com with ESMTP; 18 Jun 2020 02:21:26 -0700 From: Ting Fu To: ffmpeg-devel@ffmpeg.org Date: Thu, 18 Jun 2020 17:15:35 +0800 Message-Id: <20200618091536.8733-5-ting.fu@intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20200618091536.8733-1-ting.fu@intel.com> References: <20200618091536.8733-1-ting.fu@intel.com> Subject: [FFmpeg-devel] [PATCH 5/6] dnn_backend_native_layer_mathunary: add atan support X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" It can be tested with the model generated with below python script: import tensorflow as tf import numpy as np import imageio in_img = imageio.imread('input.jpeg') in_img = in_img.astype(np.float32)/255.0 in_data = in_img[np.newaxis, :] x = tf.placeholder(tf.float32, shape=[1, None, None, 3], name='dnn_in') x1 = tf.atan(x) x2 = tf.divide(x1, 3.1416/4) # pi/4 y = tf.identity(x2, name='dnn_out') sess=tf.Session() sess.run(tf.global_variables_initializer()) graph_def = tf.graph_util.convert_variables_to_constants(sess, sess.graph_def, ['dnn_out']) tf.train.write_graph(graph_def, '.', 'image_process.pb', as_text=False) print("image_process.pb generated, please use \ path_to_ffmpeg/tools/python/convert.py to generate image_process.model\n") output = sess.run(y, feed_dict={x: in_data}) imageio.imsave("out.jpg", np.squeeze(output)) Signed-off-by: Ting Fu --- libavfilter/dnn/dnn_backend_native_layer_mathunary.c | 4 ++++ libavfilter/dnn/dnn_backend_native_layer_mathunary.h | 1 + tools/python/convert_from_tensorflow.py | 2 +- tools/python/convert_header.py | 2 +- 4 files changed, 7 insertions(+), 2 deletions(-) diff --git a/libavfilter/dnn/dnn_backend_native_layer_mathunary.c b/libavfilter/dnn/dnn_backend_native_layer_mathunary.c index d130058546..42615c43d5 100644 --- a/libavfilter/dnn/dnn_backend_native_layer_mathunary.c +++ b/libavfilter/dnn/dnn_backend_native_layer_mathunary.c @@ -100,6 +100,10 @@ int dnn_execute_layer_math_unary(DnnOperand *operands, const int32_t *input_oper for (int i = 0; i < dims_count; ++i) dst[i] = acos(src[i]); return 0; + case DMUO_ATAN: + for (int i = 0; i < dims_count; ++i) + dst[i] = atan(src[i]); + return 0; default: return -1; } diff --git a/libavfilter/dnn/dnn_backend_native_layer_mathunary.h b/libavfilter/dnn/dnn_backend_native_layer_mathunary.h index f146248567..13fa33178a 100644 --- a/libavfilter/dnn/dnn_backend_native_layer_mathunary.h +++ b/libavfilter/dnn/dnn_backend_native_layer_mathunary.h @@ -36,6 +36,7 @@ typedef enum { DMUO_TAN = 3, DMUO_ASIN = 4, DMUO_ACOS = 5, + DMUO_ATAN = 6, DMUO_COUNT } DNNMathUnaryOperation; diff --git a/tools/python/convert_from_tensorflow.py b/tools/python/convert_from_tensorflow.py index 78297e48a9..b90c31c495 100644 --- a/tools/python/convert_from_tensorflow.py +++ b/tools/python/convert_from_tensorflow.py @@ -72,7 +72,7 @@ class TFConverter: self.conv2d_scopename_inputname_dict = {} self.op2code = {'Conv2D':1, 'DepthToSpace':2, 'MirrorPad':3, 'Maximum':4, 'MathBinary':5, 'MathUnary':6} self.mathbin2code = {'Sub':0, 'Add':1, 'Mul':2, 'RealDiv':3, 'Minimum':4} - self.mathun2code = {'Abs':0, 'Sin':1, 'Cos':2, 'Tan':3, 'Asin':4, 'Acos':5} + self.mathun2code = {'Abs':0, 'Sin':1, 'Cos':2, 'Tan':3, 'Asin':4, 'Acos':5, 'Atan':6} self.mirrorpad_mode = {'CONSTANT':0, 'REFLECT':1, 'SYMMETRIC':2} self.name_operand_dict = {} diff --git a/tools/python/convert_header.py b/tools/python/convert_header.py index 4a8e44b4aa..73cf23bf53 100644 --- a/tools/python/convert_header.py +++ b/tools/python/convert_header.py @@ -23,4 +23,4 @@ str = 'FFMPEGDNNNATIVE' major = 1 # increase minor when we don't have to re-convert the model file -minor = 11 +minor = 12