Add PP-ModNet and PP-HumanMatting Support (PaddlePaddle#240)

* first commit for yolov7 * pybind for yolov7 * CPP README.md * CPP README.md * modified yolov7.cc * README.md * python file modify * delete license in fastdeploy/ * repush the conflict part * README.md modified * README.md modified * file path modified * file path modified * file path modified * file path modified * file path modified * README modified * README modified * move some helpers to private * add examples for yolov7 * api.md modified * api.md modified * api.md modified * YOLOv7 * yolov7 release link * yolov7 release link * yolov7 release link * copyright * change some helpers to private * change variables to const and fix documents. * gitignore * Transfer some funtions to private member of class * Transfer some funtions to private member of class * Merge from develop (PaddlePaddle#9) * Fix compile problem in different python version (PaddlePaddle#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <[email protected]> * Add PaddleDetetion/PPYOLOE model support (PaddlePaddle#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (PaddlePaddle#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (PaddlePaddle#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> * first commit for yolor * for merge * Develop (PaddlePaddle#11) * Fix compile problem in different python version (PaddlePaddle#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <[email protected]> * Add PaddleDetetion/PPYOLOE model support (PaddlePaddle#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (PaddlePaddle#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (PaddlePaddle#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> * Yolor (PaddlePaddle#16) * Develop (PaddlePaddle#11) (PaddlePaddle#12) * Fix compile problem in different python version (PaddlePaddle#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <[email protected]> * Add PaddleDetetion/PPYOLOE model support (PaddlePaddle#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (PaddlePaddle#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (PaddlePaddle#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> * Develop (PaddlePaddle#13) * Fix compile problem in different python version (PaddlePaddle#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <[email protected]> * Add PaddleDetetion/PPYOLOE model support (PaddlePaddle#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (PaddlePaddle#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (PaddlePaddle#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * Develop (PaddlePaddle#14) * Fix compile problem in different python version (PaddlePaddle#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <[email protected]> * Add PaddleDetetion/PPYOLOE model support (PaddlePaddle#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (PaddlePaddle#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (PaddlePaddle#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> Co-authored-by: Jason <[email protected]> * add is_dynamic for YOLO series (PaddlePaddle#22) * modify ppmatting backend and docs * modify ppmatting docs * fix the PPMatting size problem * fix LimitShort's log * retrigger ci * modify PPMatting docs * modify the way for dealing with LimitShort * add pphumanmatting and modnet series * docs of PPMatting series * add explanation of newly added processors and fix processors * Modify LimitShort function and ppmatting.cc * modify ResizeByShort and ppmatting.cc * change resize_to_int_mult to limit_by_stride and delete resize_by_input_shape * retrigger ci * retrigger ci * fix problem produced by ResizeByShort * Update eigen.cmake * Delete eigen.cmake * refine code * add test file for ppmatting series * add squeeze for fd_tensor and modify ppmatting.cc Co-authored-by: Jason <[email protected]> Co-authored-by: root <[email protected]> Co-authored-by: DefTruth <[email protected]> Co-authored-by: huangjianhui <[email protected]> Co-authored-by: Jason <[email protected]>
TomasRan · Oct 7, 2022 · 0692dcc · 0692dcc
1 parent 1005a09
commit 0692dcc
Show file tree

Hide file tree

Showing 21 changed files with 524 additions and 133 deletions.
diff --git a/examples/vision/matting/README.md b/examples/vision/matting/README.md
@@ -6,3 +6,5 @@ FastDeploy目前支持如下抠图模型部署
 | :--- | :--- | :------- | :--- |
 | [ZHKKKe/MODNet](./modnet) | MODNet 系列模型 | ONNX | [CommitID:28165a4](https://github.com/ZHKKKe/MODNet/commit/28165a4) |
 | [PaddleSeg/PPMatting](./ppmatting) | PPMatting 系列模型 | Paddle | [Release/2.6](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting) |
+| [PaddleSeg/PPHumanMatting](./ppmatting) | PPHumanMatting 系列模型 | Paddle | [Release/2.6](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting) |
+| [PaddleSeg/ModNet](./ppmatting) | ModNet 系列模型 | Paddle | [Release/2.6](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting) |
diff --git a/examples/vision/matting/ppmatting/README.md b/examples/vision/matting/ppmatting/README.md
@@ -9,11 +9,13 @@
 目前FastDeploy支持如下模型的部署
 
 - [PPMatting系列模型](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting)
+- [PPHumanMatting系列模型](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting)
+- [ModNet系列模型](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting)
 
 
 ## 导出部署模型
 
-在部署前，需要先将PPMatting导出成部署模型，导出步骤参考文档[导出模型](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting)
+在部署前，需要先将PPMatting导出成部署模型，导出步骤参考文档[导出模型](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/Matting)(Tips:导出PPMatting系列模型和PPHumanMatting系列模型需要设置导出脚本的`--input_shape`参数)
 
 
 ## 下载预训练模型
@@ -25,8 +27,12 @@
 
 | 模型                                                               | 参数大小    | 精度    | 备注 |
 |:---------------------------------------------------------------- |:----- |:----- | :------ |
-| [PPMatting-512](https://bj.bcebos.com/paddlehub/fastdeploy/PP-Matting-512.tgz) | 87MB | - |
-| [PPMatting-1024](https://bj.bcebos.com/paddlehub/fastdeploy/PP-Matting-1024.tgz) | 87MB | - |
+| [PPMatting-512](https://bj.bcebos.com/paddlehub/fastdeploy/PP-Matting-512.tgz) | 106MB | - |
+| [PPMatting-1024](https://bj.bcebos.com/paddlehub/fastdeploy/PP-Matting-1024.tgz) | 106MB | - |
+| [PPHumanMatting](https://bj.bcebos.com/paddlehub/fastdeploy/PPHumanMatting.tgz) | 247MB | - |
+| [Modnet_ResNet50_vd](https://bj.bcebos.com/paddlehub/fastdeploy/PPModnet_ResNet50_vd.tgz) | 355MB | - |
+| [Modnet_MobileNetV2](https://bj.bcebos.com/paddlehub/fastdeploy/PPModnet_MobileNetV2.tgz) | 28MB | - |
+| [Modnet_HRNet_w18](https://bj.bcebos.com/paddlehub/fastdeploy/PPModnet_HRNet_w18.tgz) | 51MB | - |
 
 
 

diff --git a/examples/vision/matting/ppmatting/cpp/infer.cc b/examples/vision/matting/ppmatting/cpp/infer.cc
@@ -81,7 +81,7 @@ void GpuInfer(const std::string& model_dir, const std::string& image_file,
   cv::imwrite("visualized_result.jpg", vis_im_with_bg);
   cv::imwrite("visualized_result_fg.jpg", vis_im);
   std::cout << "Visualized result save in ./visualized_result_replaced_bg.jpg "
-               "and ./visualized_result_fg.jpgg"
+               "and ./visualized_result_fg.jpg"
             << std::endl;
 }
 

diff --git a/fastdeploy/core/fd_tensor.cc b/fastdeploy/core/fd_tensor.cc
@@ -85,6 +85,13 @@ void FDTensor::ExpandDim(int64_t axis) {
   shape.insert(shape.begin() + axis, 1);
 }
 
+void FDTensor::Squeeze(int64_t axis) {
+  size_t ndim = shape.size();
+  FDASSERT(axis >= 0 && axis < ndim,
+           "The allowed 'axis' must be in range of (0, %lu)!", ndim);
+  shape.erase(shape.begin() + axis);
+}
+
 void FDTensor::Allocate(const std::vector<int64_t>& new_shape,
                         const FDDataType& data_type,
                         const std::string& tensor_name,

diff --git a/fastdeploy/core/fd_tensor.h b/fastdeploy/core/fd_tensor.h
@@ -71,6 +71,10 @@ struct FASTDEPLOY_DECL FDTensor {
   // at the `axis` position in the expanded Tensor shape.
   void ExpandDim(int64_t axis = 0);
 
+  // Squeeze the shape of a Tensor. Erase the axis that will appear
+  // at the `axis` position in the squeezed Tensor shape.
+  void Squeeze(int64_t axis = 0);
+
   // Initialize Tensor
   // Include setting attribute for tensor
   // and allocate cpu memory buffer

diff --git a/fastdeploy/vision/common/processors/crop.cc b/fastdeploy/vision/common/processors/crop.cc
@@ -0,0 +1,65 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "fastdeploy/vision/common/processors/crop.h"
+
+namespace fastdeploy {
+namespace vision {
+
+bool Crop::CpuRun(Mat* mat) {
+  cv::Mat* im = mat->GetCpuMat();
+  int height = static_cast<int>(im->rows);
+  int width = static_cast<int>(im->cols);
+  if (height < height_ + offset_h_ || width < width_ + offset_w_) {
+    FDERROR << "[Crop] Cannot crop [" << height_ << ", " << width_
+            << "] from the input image [" << height << ", " << width
+            << "], with offset [" << offset_h_ << ", " << offset_w_ << "]."
+            << std::endl;
+    return false;
+  }
+  cv::Rect crop_roi(offset_w_, offset_h_, width_, height_);
+  *im = (*im)(crop_roi);
+  mat->SetWidth(width_);
+  mat->SetHeight(height_);
+  return true;
+}
+
+#ifdef ENABLE_OPENCV_CUDA
+bool Crop::GpuRun(Mat* mat) {
+  cv::cuda::GpuMat* im = mat->GetGpuMat();
+  int height = static_cast<int>(im->rows);
+  int width = static_cast<int>(im->cols);
+  if (height < height_ + offset_h_ || width < width_ + offset_w_) {
+    FDERROR << "[Crop] Cannot crop [" << height_ << ", " << width_
+            << "] from the input image [" << height << ", " << width
+            << "], with offset [" << offset_h_ << ", " << offset_w_ << "]."
+            << std::endl;
+    return false;
+  }
+  cv::Rect crop_roi(offset_w_, offset_h_, width_, height_);
+  *im = (*im)(crop_roi);
+  mat->SetWidth(width_);
+  mat->SetHeight(height_);
+  return true;
+}
+#endif
+
+bool Crop::Run(Mat* mat, int offset_w, int offset_h, int width, int height,
+               ProcLib lib) {
+  auto c = Crop(offset_w, offset_h, width, height);
+  return c(mat, lib);
+}
+
+}  // namespace vision
+}  // namespace fastdeploy
diff --git a/fastdeploy/vision/common/processors/crop.h b/fastdeploy/vision/common/processors/crop.h
@@ -0,0 +1,47 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#pragma once
+
+#include "fastdeploy/vision/common/processors/base.h"
+
+namespace fastdeploy {
+namespace vision {
+
+class Crop : public Processor {
+ public:
+  Crop(int offset_w, int offset_h, int width, int height) {
+    offset_w_ = offset_w;
+    offset_h_ = offset_h;
+    width_ = width;
+    height_ = height;
+  }
+  bool CpuRun(Mat* mat);
+#ifdef ENABLE_OPENCV_CUDA
+  bool GpuRun(Mat* mat);
+#endif
+  std::string Name() { return "Crop"; }
+
+  static bool Run(Mat* mat, int offset_w, int offset_h, int width, int height,
+                  ProcLib lib = ProcLib::OPENCV_CPU);
+
+ private:
+  int offset_w_;
+  int offset_h_;
+  int height_;
+  int width_;
+};
+
+}  // namespace vision
+}  // namespace fastdeploy
diff --git a/...n/common/processors/resize_to_int_mult.cc → ...sion/common/processors/limit_by_stride.cc b/...n/common/processors/resize_to_int_mult.cc → ...sion/common/processors/limit_by_stride.cc
@@ -12,17 +12,17 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-#include "fastdeploy/vision/common/processors/resize_to_int_mult.h"
+#include "fastdeploy/vision/common/processors/limit_by_stride.h"
 
 namespace fastdeploy {
 namespace vision {
 
-bool ResizeToIntMult::CpuRun(Mat* mat) {
+bool LimitByStride::CpuRun(Mat* mat) {
   cv::Mat* im = mat->GetCpuMat();
   int origin_w = im->cols;
   int origin_h = im->rows;
-  int rw = origin_w - origin_w % mult_int_;
-  int rh = origin_h - origin_h % mult_int_;
+  int rw = origin_w - origin_w % stride_;
+  int rh = origin_h - origin_h % stride_;
   if (rw != origin_w || rh != origin_w) {
     cv::resize(*im, *im, cv::Size(rw, rh), 0, 0, interp_);
     mat->SetWidth(im->cols);
@@ -32,13 +32,13 @@ bool ResizeToIntMult::CpuRun(Mat* mat) {
 }
 
 #ifdef ENABLE_OPENCV_CUDA
-bool ResizeToIntMult::GpuRun(Mat* mat) {
+bool LimitByStride::GpuRun(Mat* mat) {
   cv::cuda::GpuMat* im = mat->GetGpuMat();
   int origin_w = im->cols;
   int origin_h = im->rows;
   im->convertTo(*im, CV_32FC(im->channels()));
-  int rw = origin_w - origin_w % mult_int_;
-  int rh = origin_h - origin_h % mult_int_;
+  int rw = origin_w - origin_w % stride_;
+  int rh = origin_h - origin_h % stride_;
   if (rw != origin_w || rh != origin_w) {
     cv::cuda::resize(*im, *im, cv::Size(rw, rh), 0, 0, interp_);
     mat->SetWidth(im->cols);
@@ -48,8 +48,8 @@ bool ResizeToIntMult::GpuRun(Mat* mat) {
 }
 #endif
 
-bool ResizeToIntMult::Run(Mat* mat, int mult_int, int interp, ProcLib lib) {
-  auto r = ResizeToIntMult(mult_int, interp);
+bool LimitByStride::Run(Mat* mat, int stride, int interp, ProcLib lib) {
+  auto r = LimitByStride(stride, interp);
   return r(mat, lib);
 }
 }  // namespace vision

diff --git a/...on/common/processors/resize_to_int_mult.h → ...ision/common/processors/limit_by_stride.h b/...on/common/processors/resize_to_int_mult.h → ...ision/common/processors/limit_by_stride.h
@@ -19,24 +19,27 @@
 namespace fastdeploy {
 namespace vision {
 
-class ResizeToIntMult : public Processor {
+class LimitByStride : public Processor {
  public:
-  explicit ResizeToIntMult(int mult_int = 32, int interp = 1) {
-    mult_int_ = mult_int;
+  explicit LimitByStride(int stride = 32, int interp = 1) {
+    stride_ = stride;
     interp_ = interp;
   }
+
+  // Resize Mat* mat to make the size divisible by stride_.
+
   bool CpuRun(Mat* mat);
 #ifdef ENABLE_OPENCV_CUDA
   bool GpuRun(Mat* mat);
 #endif
-  std::string Name() { return "ResizeToIntMult"; }
+  std::string Name() { return "LimitByStride"; }
 
-  static bool Run(Mat* mat, int mult_int = 32, int interp = 1,
+  static bool Run(Mat* mat, int stride = 32, int interp = 1,
                   ProcLib lib = ProcLib::OPENCV_CPU);
 
  private:
   int interp_;
-  int mult_int_;
+  int stride_;
 };
 }  // namespace vision
 }  // namespace fastdeploy
diff --git a/fastdeploy/vision/common/processors/limit_long.cc b/fastdeploy/vision/common/processors/limit_long.cc
@@ -0,0 +1,70 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "fastdeploy/vision/common/processors/limit_long.h"
+
+namespace fastdeploy {
+namespace vision {
+
+bool LimitLong::CpuRun(Mat* mat) {
+  cv::Mat* im = mat->GetCpuMat();
+  int origin_w = im->cols;
+  int origin_h = im->rows;
+  int im_size_max = std::max(origin_w, origin_h);
+  int target = im_size_max;
+  if (max_long_ > 0 && im_size_max > max_long_) {
+    target = max_long_;
+  } else if (min_long_ > 0 && im_size_max < min_long_) {
+    target = min_long_;
+  }
+  if (target != im_size_max) {
+    double scale =
+        static_cast<double>(target) / static_cast<double>(im_size_max);
+    cv::resize(*im, *im, cv::Size(), scale, scale, interp_);
+    mat->SetWidth(im->cols);
+    mat->SetHeight(im->rows);
+  }
+  return true;
+}
+
+#ifdef ENABLE_OPENCV_CUDA
+bool LimitLong::GpuRun(Mat* mat) {
+  cv::cuda::GpuMat* im = mat->GetGpuMat();
+  int origin_w = im->cols;
+  int origin_h = im->rows;
+  im->convertTo(*im, CV_32FC(im->channels()));
+  int im_size_max = std::max(origin_w, origin_h);
+  int target = im_size_max;
+  if (max_long_ > 0 && im_size_max > max_long_) {
+    target = max_long_;
+  } else if (min_long_ > 0 && im_size_max < min_long_) {
+    target = min_long_;
+  }
+  if (target != im_size_max) {
+    double scale =
+        static_cast<double>(target) / static_cast<double>(im_size_max);
+    cv::cuda::resize(*im, *im, cv::Size(), scale, scale, interp_);
+    mat->SetWidth(im->cols);
+    mat->SetHeight(im->rows);
+  }
+  return true;
+}
+#endif
+
+bool LimitLong::Run(Mat* mat, int max_long, int min_long, ProcLib lib) {
+  auto l = LimitLong(max_long, min_long);
+  return l(mat, lib);
+}
+}  // namespace vision
+}  // namespace fastdeploy
diff --git a/fastdeploy/vision/common/processors/limit_long.h b/fastdeploy/vision/common/processors/limit_long.h
@@ -0,0 +1,51 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#pragma once
+
+#include "fastdeploy/vision/common/processors/base.h"
+
+namespace fastdeploy {
+namespace vision {
+
+class LimitLong : public Processor {
+ public:
+  explicit LimitLong(int max_long = -1, int min_long = -1, int interp = 1) {
+    max_long_ = max_long;
+    min_long_ = min_long;
+    interp_ = interp;
+  }
+
+  // Limit the long edge of image.
+  // If the long edge is larger than max_long_, resize the long edge
+  // to max_long_, while scale the short edge proportionally.
+  // If the long edge is smaller than min_long_, resize the long edge
+  // to min_long_, while scale the short edge proportionally.
+  bool CpuRun(Mat* mat);
+#ifdef ENABLE_OPENCV_CUDA
+  bool GpuRun(Mat* mat);
+#endif
+  std::string Name() { return "LimitLong"; }
+
+  static bool Run(Mat* mat, int max_long = -1, int min_long = -1,
+                  ProcLib lib = ProcLib::OPENCV_CPU);
+  int GetMaxLong() const { return max_long_; }
+
+ private:
+  int max_long_;
+  int min_long_;
+  int interp_;
+};
+}  // namespace vision
+}  // namespace fastdeploy