Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Support all resize types in kp model #275

Merged
merged 8 commits into from
Mar 7, 2025
Merged

Support all resize types in kp model #275

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
2528127
Support all resize types in kp model
sovrasov Mar 4, 2025
60c0a26
Update cpp implementation
sovrasov Mar 5, 2025
4c7ab31
Update kp ref
sovrasov Mar 5, 2025
37def73
Update scores computation
sovrasov Mar 5, 2025
2d91938
Update ref scores
sovrasov Mar 5, 2025
0e3f454
Add visibility score computation to cpp
sovrasov Mar 5, 2025
63d0e44
Fix a typo
sovrasov Mar 5, 2025
18f124c
Skip direct KP model inference in cpp
sovrasov Mar 7, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 19 additions & 3 deletions src/cpp/models/src/keypoint_detection.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,7 @@ void colArgMax(const cv::Mat& src,
DetectedKeypoints decode_simcc(const cv::Mat& simcc_x,
const cv::Mat& simcc_y,
const cv::Point2f& extra_scale = cv::Point2f(1.f, 1.f),
const cv::Point2i& extra_offset = cv::Point2f(0.f, 0.f),
bool apply_softmax = false,
float simcc_split_ratio = 2.0f) {
cv::Mat x_locs, max_val_x;
Expand All @@ -64,8 +65,9 @@ DetectedKeypoints decode_simcc(const cv::Mat& simcc_x,
std::vector<cv::Point2f> keypoints(x_locs.rows);
cv::Mat scores = cv::Mat::zeros(x_locs.rows, 1, CV_32F);
for (int i = 0; i < x_locs.rows; i++) {
keypoints[i] =
cv::Point2f(x_locs.at<int>(i) * extra_scale.x, y_locs.at<int>(i) * extra_scale.y) / simcc_split_ratio;
keypoints[i] = cv::Point2f((x_locs.at<int>(i) - extra_offset.x) * extra_scale.x,
(y_locs.at<int>(i) - extra_offset.y) * extra_scale.y) /
simcc_split_ratio;
scores.at<float>(i) = std::min(max_val_x.at<float>(i), max_val_y.at<float>(i));

if (scores.at<float>(i) <= 0.f) {
Expand Down Expand Up @@ -220,8 +222,22 @@ std::unique_ptr<ResultBase> KeypointDetectionModel::postprocess(InferenceResult&
float inverted_scale_x = static_cast<float>(image_data.inputImgWidth) / netInputWidth,
inverted_scale_y = static_cast<float>(image_data.inputImgHeight) / netInputHeight;

int pad_left = 0, pad_top = 0;
if (RESIZE_KEEP_ASPECT == resizeMode || RESIZE_KEEP_ASPECT_LETTERBOX == resizeMode) {
inverted_scale_x = inverted_scale_y = std::max(inverted_scale_x, inverted_scale_y);
if (RESIZE_KEEP_ASPECT_LETTERBOX == resizeMode) {
pad_left = (netInputWidth -
static_cast<int>(std::round(static_cast<float>(image_data.inputImgWidth) / inverted_scale_x))) /
2;
pad_top = (netInputHeight -
static_cast<int>(std::round(static_cast<float>(image_data.inputImgHeight) / inverted_scale_y))) /
2;
}
}

result->poses.emplace_back(
decode_simcc(pred_x_mat, pred_y_mat, {inverted_scale_x, inverted_scale_y}, apply_softmax));
decode_simcc(pred_x_mat, pred_y_mat, {inverted_scale_x, inverted_scale_y}, {pad_left, pad_top}, apply_softmax));

return std::unique_ptr<ResultBase>(result);
}

Expand Down
14 changes: 13 additions & 1 deletion src/python/model_api/models/keypoint_detection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,19 @@ def postprocess(
orig_h, orig_w = meta["original_shape"][:2]
kp_scale_h = orig_h / self.h
kp_scale_w = orig_w / self.w
batch_keypoints = batch_keypoints.squeeze() * np.array([kp_scale_w, kp_scale_h])

batch_keypoints = batch_keypoints.squeeze()

if self.resize_type in ["fit_to_window", "fit_to_window_letterbox"]:
inverted_scale = max(kp_scale_h, kp_scale_w)
kp_scale_h = kp_scale_w = inverted_scale
if self.resize_type == "fit_to_window_letterbox":
pad_left = (self.w - round(orig_w / inverted_scale)) // 2
pad_top = (self.h - round(orig_h / inverted_scale)) // 2
batch_keypoints -= np.array([pad_left, pad_top])

batch_keypoints *= np.array([kp_scale_w, kp_scale_h])

return DetectedKeypoints(batch_keypoints, batch_scores.squeeze())

@classmethod
Expand Down