Sparse optical flow via Lucas-Kanade tracker. More...

#include "Features.hpp"

Include dependency graph for OpticalFlow.hpp:

This graph shows which files directly or indirectly include this file:

Classes
struct	MayaFlux::Kinesis::Vision::TrackResult
	Result for a single tracked point. More...

Namespaces
namespace	MayaFlux
	Main namespace for the Maya Flux audio engine.

namespace	MayaFlux::Kinesis

namespace	MayaFlux::Kinesis::Vision

Functions
std::vector< TrackResult >	MayaFlux::Kinesis::Vision::track_keypoints (std::span< const float > prev_gray, std::span< const float > curr_gray, std::span< const float > grad_dx, std::span< const float > grad_dy, uint32_t w, uint32_t h, std::span< const glm::vec2 > prev_points, uint32_t window_radius=7, uint32_t max_iterations=20, float eigen_threshold=1e-4F, float error_threshold=0.3F)
	Track keypoints using a pre-computed Sobel gradient of prev_gray.

std::vector< TrackResult >	MayaFlux::Kinesis::Vision::track_keypoints (std::span< const float > prev_gray, std::span< const float > curr_gray, uint32_t w, uint32_t h, std::span< const glm::vec2 > prev_points, uint32_t window_radius=7, uint32_t max_iterations=20, float eigen_threshold=1e-4F, float error_threshold=0.3F)
	Track keypoints from prev_gray to curr_gray via Lucas-Kanade.

Detailed Description

Sparse optical flow via Lucas-Kanade tracker.

Pure functions. No MayaFlux type dependencies.

Conventions

All inputs are single-channel normalised float in [0, 1]
w and h are pixel dimensions; caller ensures span size == w * h
Keypoint positions are normalised to [0, 1]
window_radius is the half-size of the tracking window in pixels
Points that fail to track are returned at their previous position with tracked = false
Parallelism handled internally via Parallel::par_unseq

Definition in file OpticalFlow.hpp.