|
override string | getDefaultName () |
|
void | detect (Mat image, MatOfKeyPoint keypoints, Mat mask) |
| Detects keypoints in an image (first variant) or image set (second variant). More...
|
|
void | detect (Mat image, MatOfKeyPoint keypoints) |
| Detects keypoints in an image (first variant) or image set (second variant). More...
|
|
void | detect (List< Mat > images, List< MatOfKeyPoint > keypoints, List< Mat > masks) |
|
void | detect (List< Mat > images, List< MatOfKeyPoint > keypoints) |
|
void | compute (Mat image, MatOfKeyPoint keypoints, Mat descriptors) |
| Computes the descriptors for a set of keypoints detected in an image (first variant) or image set (second variant). More...
|
|
void | compute (List< Mat > images, List< MatOfKeyPoint > keypoints, List< Mat > descriptors) |
|
void | detectAndCompute (Mat image, Mat mask, MatOfKeyPoint keypoints, Mat descriptors, bool useProvidedKeypoints) |
|
void | detectAndCompute (Mat image, Mat mask, MatOfKeyPoint keypoints, Mat descriptors) |
|
int | descriptorSize () |
|
int | descriptorType () |
|
int | defaultNorm () |
|
void | write (string fileName) |
|
void | read (string fileName) |
|
override bool | empty () |
| Returns true if the Algorithm is empty (e.g. in the very beginning or after unsuccessful read. More...
|
|
IntPtr | getNativeObjAddr () |
|
virtual void | clear () |
| Clears the algorithm state. More...
|
|
void | save (string filename) |
|
void | Dispose () |
|
void | ThrowIfDisposed () |
|
Class implementing TEBLID (Triplet-based Efficient Binary Local Image Descriptor), described in [Suarez2021TEBLID].
TEBLID stands for Triplet-based Efficient Binary Local Image Descriptor, although originally it was called BAD [Suarez2021TEBLID]. It is an improvement over BEBLID [Suarez2020BEBLID], that uses triplet loss, hard negative mining, and anchor swap to improve the image matching results. It is able to describe keypoints from any detector just by changing the scale_factor parameter. TEBLID is as efficient as ORB, BEBLID or BRISK, but the triplet-based training objective selected more discriminative features that explain the accuracy gain. It is also more compact than BEBLID, when running the AKAZE example with 10000 keypoints detected by ORB, BEBLID obtains 561 inliers (75%) with 512 bits, whereas TEBLID obtains 621 (75.2%) with 256 bits. ORB obtains only 493 inliers (63%).
If you find this code useful, please add a reference to the following paper: <BLOCKQUOTE> Iago Suárez, José M. Buenaposada, and Luis Baumela. Revisiting Binary Local Image Description for Resource Limited Devices. IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 8317-8324, Oct. 2021. </BLOCKQUOTE>
The descriptor was trained in Liberty split of the UBC datasets [winder2007learning] .