2021-12-07 10:57:55 +08:00
|
|
|
// Copyright (c) OpenMMLab. All rights reserved.
|
|
|
|
|
|
|
|
#include "crop.h"
|
|
|
|
|
|
|
|
#include "archive/json_archive.h"
|
|
|
|
|
|
|
|
using namespace std;
|
|
|
|
|
|
|
|
namespace mmdeploy {
|
|
|
|
|
|
|
|
CenterCropImpl::CenterCropImpl(const Value& args) : TransformImpl(args) {
|
|
|
|
if (!args.contains(("crop_size"))) {
|
|
|
|
throw std::invalid_argument("'crop_size' is expected");
|
|
|
|
}
|
|
|
|
if (args["crop_size"].is_number_integer()) {
|
|
|
|
int crop_size = args["crop_size"].get<int>();
|
|
|
|
arg_.crop_size[0] = arg_.crop_size[1] = crop_size;
|
|
|
|
} else if (args["crop_size"].is_array() && args["crop_size"].size() == 2) {
|
|
|
|
arg_.crop_size[0] = args["crop_size"][0].get<int>();
|
|
|
|
arg_.crop_size[1] = args["crop_size"][1].get<int>();
|
|
|
|
} else {
|
|
|
|
throw std::invalid_argument("'crop_size' should be integer or an int array of size 2");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
Result<Value> CenterCropImpl::Process(const Value& input) {
|
2022-02-24 20:08:44 +08:00
|
|
|
MMDEPLOY_DEBUG("input: {}", to_json(input).dump(2));
|
2021-12-07 10:57:55 +08:00
|
|
|
auto img_fields = GetImageFields(input);
|
|
|
|
|
|
|
|
// copy input data, and update its properties
|
|
|
|
Value output = input;
|
|
|
|
|
|
|
|
for (auto& key : img_fields) {
|
|
|
|
auto tensor = input[key].get<Tensor>();
|
|
|
|
auto desc = tensor.desc();
|
|
|
|
int h = desc.shape[1];
|
|
|
|
int w = desc.shape[2];
|
|
|
|
int crop_height = arg_.crop_size[0];
|
|
|
|
int crop_width = arg_.crop_size[1];
|
|
|
|
|
|
|
|
int y1 = std::max(0, int(std::round((h - crop_height) / 2.0)));
|
|
|
|
int x1 = std::max(0, int(std::round((w - crop_width) / 2.0)));
|
|
|
|
int y2 = std::min(h, y1 + crop_height) - 1;
|
|
|
|
int x2 = std::min(w, x1 + crop_width) - 1;
|
|
|
|
|
|
|
|
OUTCOME_TRY(auto dst_tensor, CropImage(tensor, y1, x1, y2, x2));
|
|
|
|
|
|
|
|
auto& shape = dst_tensor.desc().shape;
|
|
|
|
|
|
|
|
output[key] = dst_tensor;
|
|
|
|
output["img_shape"] = {shape[0], shape[1], shape[2], shape[3]};
|
|
|
|
if (input.contains("scale_factor")) {
|
|
|
|
// image has been processed by `Resize` transform before.
|
|
|
|
// Compute cropped image's offset against the original image
|
|
|
|
assert(input["scale_factor"].is_array() && input["scale_factor"].size() >= 2);
|
|
|
|
float w_scale = input["scale_factor"][0].get<float>();
|
|
|
|
float h_scale = input["scale_factor"][1].get<float>();
|
|
|
|
output["offset"].push_back(x1 / w_scale);
|
|
|
|
output["offset"].push_back(y1 / h_scale);
|
|
|
|
} else {
|
|
|
|
output["offset"].push_back(x1);
|
|
|
|
output["offset"].push_back(y1);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-02-24 20:08:44 +08:00
|
|
|
MMDEPLOY_DEBUG("output: {}", to_json(output).dump(2));
|
2021-12-07 10:57:55 +08:00
|
|
|
return output;
|
|
|
|
}
|
|
|
|
|
|
|
|
CenterCrop::CenterCrop(const Value& args, int version) : Transform(args) {
|
|
|
|
auto impl_creator = Registry<CenterCropImpl>::Get().GetCreator(specified_platform_, version);
|
|
|
|
if (nullptr == impl_creator) {
|
2022-02-24 20:08:44 +08:00
|
|
|
MMDEPLOY_ERROR("'CenterCrop' is not supported on '{}' platform", specified_platform_);
|
2021-12-07 10:57:55 +08:00
|
|
|
throw std::domain_error("'Resize' is not supported on specified platform");
|
|
|
|
}
|
|
|
|
impl_ = impl_creator->Create(args);
|
|
|
|
}
|
|
|
|
|
|
|
|
class CenterCropCreator : public Creator<Transform> {
|
|
|
|
public:
|
|
|
|
const char* GetName(void) const override { return "CenterCrop"; }
|
|
|
|
int GetVersion(void) const override { return version_; }
|
|
|
|
ReturnType Create(const Value& args) override { return make_unique<CenterCrop>(args, version_); }
|
|
|
|
|
|
|
|
private:
|
|
|
|
int version_{1};
|
|
|
|
};
|
|
|
|
|
|
|
|
REGISTER_MODULE(Transform, CenterCropCreator);
|
2022-02-24 20:08:44 +08:00
|
|
|
MMDEPLOY_DEFINE_REGISTRY(CenterCropImpl);
|
2021-12-07 10:57:55 +08:00
|
|
|
} // namespace mmdeploy
|