Functions

br_about

function definition:
```
const char *br_about()
```
parameters: None
output: (const char *) Returns a string describing OpenBR
see: br_version

br_cat

Concatenates a list of galleries into 1 gallery.

function definition:

void br_cat(int num_input_galleries, const char *input_galleries[], const char *output_gallery)

parameters:

Parameter	Type	Description
num_input_galleries	int	Size of input_galleries
input_galleries[]	const char *	List of galleries
output_gallery	const char *	Pointer to store concatenated gallery

output: void
see: Cat

br_deduplicate

Removes duplicate templates in a gallery. If a galley contains n duplicates, the first n-1 duplicates in the gallery will be removed and the nth will be kept. Users are encouraged to use binary gallery formats as the entire gallery is read into memory in one call to Gallery::read.

function definition:

void br_deduplicate(const char *input_gallery, const char *output_gallery, const char *threshold)

parameters:

Parameter	Type	Description
input_gallery	const char *	Gallery to be deduplicated
output_gallery	const char *	Deduplicated gallery
threshold	const char *	Comparisons with a match score >= this value are designated to be duplicates.

output: (void)

br_cluster

Clusters one or more similarity matrices into a list of subjects. A similarity matrix is a type of Output. The current clustering algorithm is a simplified implementation of the algorithm proposed by Zhu et al¹.

function definition:

void br_cluster(int num_simmats, const char *simmats[], float aggressiveness, const char *csv)

parameters:

Parameter	Type	Description
num_simmats	int	Size of simmats
simmats[]	const char *	Array of simmat composing one large self-similarity matrix arranged in row major order.
aggressiveness	float	The higher the aggressiveness the larger the clusters. Suggested range is [0,10]
csv	const char *	The cluster results file to generate. Results are stored one row per cluster and use gallery indices.

output: (void)

br_combine_masks

Combines several equal-sized mask matrices. A comparison may not be simultaneously indentified as both a genuine and an imposter by different input masks.

function definition:

void br_combine_masks(int num_input_masks, const char input_masks[], const char output_mask, const char *method)

parameters:

Parameter	Type	Description
num_input_masks	int	Size of input_masks
input_masks[]	const char *	Array of mask matrices to combine. All matrices must have the same dimensions.
output_mask	const char *	The file to contain the resulting mask matrix
method	const char *	Possible values are: And - Ignore comparison if any input masks ignore. Or - Ignore comparison if all input masks ignore.

see: br_make_mask

br_compare

Compares each Template in the query Gallery to each Template in the target Gallery.

function definition:

void br_compare(const char *target_gallery, const char *query_gallery, const char *output = "")

parameters:

Parameter	Type	Description
target_gallery	const char *	target_gallery The Gallery file whose templates make up the columns of the output.
query_gallery	const char *	The Gallery file whose templates make up the rows of the output. A value of '.' reuses the target gallery as the query gallery.
output	const char *	(Optional) The Output file to contain the results of comparing the templates. The default behavior is to print scores to the terminal.

output: (void)
see: br_enroll

br_compare_n

Convenience function for comparing to multiple targets.

function definition:

void br_compare_n(int num_targets, const char *target_galleries[], const char *query_gallery, const char *output)

parameters:

Parameter	Type	Description
num_targets	int	Size of target_galleries
target_galleries[]	const char *	Target galleries to compare against
query_gallery	const char *	query gallery for comparison.
output	const char *	(Optional) Output file to contain the results of comparing the templates. The default behavior is to print scores to the terminal.

output: (void)
see: br_compare

br_pairwise_compare

DOCUMENT ME!

function definition:

void br_pairwise_compare(const char *target_gallery, const char *query_gallery, const char *output = "")

parameters:

Parameter Type Description

target_gallery const char * DOCUMENT ME

query_gallery const char * DOCUMENT ME

output const char * DOCUMENT ME
output: (void)

br_convert

Convert a file to a different type. Files can only be converted to types within the same group. For example formats can only be converted to other formats.

function definition:

void br_convert(const char *file_type, const char *input_file, const char *output_file)

parameters:

Parameter	Type	Description
file_type	const char *	Type of file to convert. Options are Format, Gallery or Output.
input_file	const char *	File to convert.
output_file	const char *	Output file. Type is determined by the file extension.

output: (void)

br_enroll

Constructs Template(s) from an input.

function definition:

void br_enroll(const char *input, const char *gallery = "")

parameters:

Parameter	Type	Description
input	const char *	The format or gallery to enroll.
gallery	const char *	(Optional) The Gallery file to contain the enrolled templates. By default the gallery will be held in memory and input can used as a gallery in br_compare

output: (void)
see: br_enroll_n

br_enroll_n

Convenience function for enrolling multiple inputs.

function definition:

void br_enroll_n(int num_inputs, const char *inputs[], const char *gallery = "")

parameters:

Parameter	Type	Description
num_inputs	int	Size of inputs.
inputs[]	const char *	Array of inputs to enroll.
gallery	const char *	(Optional) The Gallery file to contain the enroll templates.

output: (void)
see: br_enroll

br_project

A naive alternative to br_enroll.

function definition:

void br_project(const char *input, const char *output)

parameters:

Parameter	Type	Description
input	const char *	The format or gallery to enroll.
output	const char *	The Gallery file to contain the enrolled templates. By default the gallery will be held in memory and input can used as a gallery in br_compare

output: (void)
see: br_enroll

br_eval

Creates a .csv file containing performance metrics from evaluating the similarity matrix using the mask matrix.

function defintion:

float br_eval(const char *simmat, const char *mask, const char *csv = "", int matches = 0)

parameters:

Parameter	Type	Description
simmat	const char *	The simmat to use
mask	const char *	The mask to use.
csv	const char *	(Optional) The .csv file to contain performance metrics.
matches	int	(Optional) An integer number of matches to output around the EER. Default is 0.

output: (float) Returns the true accept rate (TAR) at a false accept rate (FAR) of one in one thousand
see: br_plot

br_assert_eval

Evaluates the similarity matrix using the mask matrix. Function aborts if TAR @ FAR = 0.001 does not meet an expected performance value.

function definition:

void br_assert_eval(const char *simmat, const char *mask, const float accuracy)

parameters:

Parameter Type Description

simmat const char * The simmat to use

mask const char * The mask

accuracy const float Desired true accept rate at false accept rate of one in one thousand.
output: (void)

br_inplace_eval

Creates a .csv file containing performance metrics from evaluating the similarity matrix using galleries containing ground truth labels.

function definition:

float br_inplace_eval(const char * simmat, const char *target, const char *query, const char *csv = "")

parameters:

Parameter	Type	Description
simmat	const char *	The simmat
target	const char *	The name of a gallery containing metadata for the target set.
query	const char *	The name of a gallery containing metadata for the query set.
csv	const char *	(Optional) The .csv file to contain performance metrics.

output: (float) Returns the true accept rate (TAR) at a false accept rate (FAR) of one in one thousand
see: br_plot

br_eval_classification

Evaluates and prints classification accuracy to terminal.

function definition:

void br_eval_classification(const char *predicted_gallery, const char *truth_gallery, const char *predicted_property = "", const char *truth_property = "")

parameters:

Parameter	Type	Description
predicted_gallery	const char *	The predicted Gallery.
truth_gallery	const char *	The ground truth Gallery.
predicted_property	const char *	(Optional) Which metadata key to use from the predicted_gallery.
truth_property	const char *	(Optional) Which metadata key to use from the truth_gallery.

output: (void)

br_eval_clustering

Evaluates and prints clustering accuracy to the terminal.

function definition:

void br_eval_clustering(const char *csv, const char *gallery, const char * truth_property)

parameters:

Parameter	Type	Description
csv	const char *	The cluster results file.
gallery	const char *	The Gallery used to generate the simmat that was clustered.
truth_property	const char *	(Optional) which metadata key to use from gallery, defaults to Label

output: (void)

br_eval_detection

Evaluates and prints detection accuracy to terminal.

function definition:

float br_eval_detection(const char *predicted_gallery, const char *truth_gallery, const char *csv = "", bool normalize = false, int minSize = 0, int maxSize = 0)

parameters:

Parameter	Type	Description
predicted_gallery	const char *	The predicted Gallery.
truth_gallery	const char *	The ground truth Gallery.
csv	const char *	(Optional) The .csv file to contain performance metrics.
normalize	bool	(Optional) Flag to normalize predicted bounding boxes for improved detection. Defaults to false.
minSize	int	(Optional) Minimum size of faces to be considered in the evaluation. Size is applied to predicted and ground truth galleries. Defaults to -1 (no minimum size).
maxSize	int	(Optional) Maximum size if faces to be considered in the evaluation. Size is applied to predicted and ground truth galleries. Defaults to -1 (no maximum size).

output: (float) Returns the true accept rate (TAR) at a false accept rate (FAR) of one in one thousand

br_eval_landmarking

Evaluates and prints landmarking accuracy to terminal.

function definition:

float br_eval_landmarking(const char *predicted_gallery, const char *truth_gallery, const char *csv = "", int normalization_index_a = 0, int normalization_index_b = 1, int sample_index = 0, int total_examples = 5)

parameters:

Parameter	Type	Description
predicted_gallery	const char *	The predicted Gallery.
truth_gallery	const char *	The ground truth Gallery.
csv	const char *	(Optional) The .csv file to contain performance metrics.
normalization_index_a	int	(Optional) The first index in the list of points to use for normalization. Default is 0.
normalization_index_b	int	(Optional) The second index in the list of points to use for normalization. Default is 1.
sample_index	int	(Optional) The index for sample landmark image in ground truth gallery. Default = 0.
total_examples	int	(Optional) The number of accurate and inaccurate examples to display. Default is 5.

output: (float) Returns the true accept rate (TAR) at a false accept rate (FAR) of one in one thousand

br_eval_regression

Evaluates regression accuracy to disk.

function definition:

void br_eval_regression(const char *predicted_gallery, const char *truth_gallery, const char *predicted_property = "", const char *truth_property = "")

parameters:

Parameter	Type	Description
predicted_gallery	const char *	The predicted Gallery
truth_gallery	const char *	The ground truth Gallery
predicted_property	const char *	(Optional) Which metadata key to use from predicted_gallery.
truth_property	const char *	(Optional) Which metadata key to use from truth_gallery.

output: (void)

br_fuse

Perform score level fusion on similarity matrices.

function definition:

void br_fuse(int num_input_simmats, const char *input_simmats[], const char *normalization, const char *fusion, const char *output_simmat)

parameters:

Parameter	Type	Description
num_input_simmats	int	Size of input_simmats.
input_simmats[]	const char *	Array of simmats. All simmats must have the same dimensions.
normalization	const char *	Valid options are: None - No score normalization. MinMax - Scores normalized to [0,1]. ZScore - Scores normalized to a standard normal curve.
fusion	const char *	Valid options are: Min - Uses the minimum score. Max - Uses the maximum score. Sum - Sums the scores. Sums can also be weighted: `SumW1:W2:...:Wn`. Replace - Replaces scores in the first matrix with scores in the second matrix when the mask is set.
output_simmat	const char *	Simmat to contain the fused scores.

output: (void)

br_initialize

Initializes the Context. Required at the beginning of any OpenBR program.

function definition:

void br_initialize(int &argc, char *argv[], const char *sdk_path = "", bool use_gui = false)

parameters:

Parameter	Type	Description
argc	int	Number of command line arguments.
argv[]	char *	Array of command line arguments.
sdk_path	const char *	(Optional) Path to the OpenBR sdk. If no path is provided OpenBR will try and find the sdk automatically.
use_gui	bool	(Optional) Enable OpenBR to use make GUI windows. Default is false.

output: (void)
see: br_finalize

br_initialize_default

Initializes the Context with default arguments.

function definition:
```
void br_initialize_default()
```
parameters: None
output: (void)
see: br_finalize

br_finalize

Finalizes the context. Required at the end of any OpenBR program.

function definition:
```
void br_finalize()
```
parameters: None
output: (void)
see: br_initialize

br_is_classifier

Checks if the provided algorithm is a classifier. Wrapper of IsClassifier.

function definition:

bool br_is_classifier(const char *algorithm)

parameters:

Parameter Type Description

algorithm const char * Algorithm to check.
output: (bool) Returns true if the algorithm is a classifier (does not have an associated distance)
see: IsClassifier

br_make_mask

Constructs a mask from target and query inputs.

function definition:

void br_make_mask(const char *target_input, const char *query_input, const char *mask)

parameters:

Parameter Type Description

target_input const char * The target Gallery

query_input const char * The query Gallery

mask const char * The file to contain the resulting mask.
output: (void)
see: br_combine_masks

br_make_pairwise_mask

Constructs a mask from target and query inputs considering the target and input sets to be definite pairwise comparisons.

function definition:

void br_make_pairwise_mask(const char *target_input, const char *query_input, const char *mask)

parameters:

Parameter	Type	Description
target_input	const char *	The target Gallery
query_input	const char *	The query Gallery
mask	const char *	The file to contain the resulting mask.

output: (void)
see: br_combine_masks

br_most_recent_message

Returns the most recent line sent to stderr. For information on input string buffers please look here

function definition:

int br_most_recent_message(char * buffer, int buffer_length)

parameters:

Parameter Type Description

buffer char * Buffer to store the last line in.

buffer_length int Length of the buffer.
output: (int) Returns the required size of the input buffer for the most recent message to fit completely
see: br_progress, br_time_remaining

br_objects

Returns names and parameters for the requested objects. Each object is newline seperated. Arguments are seperated from the object name with a tab. This function uses QRegExp syntax.

function definition:

int br_objects(char * buffer, int buffer_length, const char *abstractions = ".*", const char *implementations = ".*", bool parameters = true)

parameters:

Parameter	Type	Description
buffer	char *	Output buffer for results.
buffer_length	int	Length of output buffer.
abstractions	const char *	(Optional) Regular expression of the abstractions to search. Default is ".*".
implementations	const char *	(Optional) Regular expression of the implementations to search. Default is ".*".
parameters	bool	(Optional) Include parameters after object name. Default is true.

output: (int) Returns the required size of the input buffer for the returned objects to fit completely

br_plot

Renders recognition performance figures for a set of .csv files created by br_eval.

In order of their output, the figures are: 1. Metadata table 2. Receiver Operating Characteristic (ROC) 3. Detection Error Tradeoff (DET) 4. Identification Error Tradeoff (IET) 5. Cumulative Match Characteristic (CMC) 6. Score Distribution (SD) histogram 7. True Accept Rate Bar Chart (BC) 8. Error Rate (ERR) curve

Two files will be created: * destination.R which is the auto-generated R script used to render the figures. * destination.pdf which has all of the figures in one file multi-page file (note that destination%1d.png will output each figure to a separate file).

OpenBR uses file and folder names to automatically determine the plot legend. For example, let's consider the case where three algorithms (A, B, & C) were each evaluated on two datasets (Y & Z). The suggested way to plot these experiments on the same graph is to create a folder named Algorithm_Dataset that contains the six .csv files produced by br_eval A_Y.csv, A_Z.csv, B_Y.csv, B_Z.csv, C_Y.csv, & C_Z.csv. The '_' character plays a special role in determining the legend title(s) and value(s). In this case, A, B, & C will be identified as different values of type Algorithm, and each will be assigned its own color; Y & Z will be identified as different values of type Dataset, and each will be assigned its own line style. Matches around the EER will be displayed if the matches parameter is set in br_eval.

It is possible to customize some aspects of your plots using the File key/value metadata convention; possible keys are described below.

Key	Value	Description
smooth	QString	The file pivot to average across evaluation splits. Typically "Dataset" if using the folder name from above.
ncol	int	Number of columns in plot legends.
confidence	float	Confidence interval calculated for smooth, defaults to `0.95`
metadata	bool	Optional plot metadata, defaults to `true`
csv	bool	Optional output metadata tables to csv, defaults to `false`
*Options	QStringList	Key/value list of options for a specific plot. Plots include "roc", "det", "iet", "cmc"

Specific plot options are described below:

Key	Value	Description
title	QString	Plot title
size	float	Line width
legendPosition	QPointF	Legend coordinates on plot, ex. legendPosition=(X,Y)
textSize	float	Size of text for title, legend and axes
xTitle/yTitle	QString	Title for x/y axis
xLog/yLog	bool	Plot log scale for x/y axis
xLimits/yLimits	QPointF	Set x/y axis limits, ex. xLimits=(lower,upper)
xLabels/yLabels	QString	Labels for ticks on x/y axis, ex. xLabeles=percent or xLabels=(1,5,10)
xBreaks/yBreaks	QString	Specify breaks/ticks on x/y axis, ex. xBreaks=pretty_breaks(n=10) or xBreaks=(1,5,10)

If specifying plot options it is a good idea to wrap the destination file in single quotes to avoid parsing errors. The example below plots plots the six br_eval results in the Algorithm_Dataset folder described above, sets the number of legend columns and specifies some options for the CMC plot.

br -plot Algorithm_Dataset/* 'destination.pdf[ncol=3,cmcOptions=[xLog=false,xLimits=(1,20),xBreaks=pretty_breaks(n=10),xTitle=Ranks 1 through 20]]'

This function requires a current R installation with the following packages:

    install.packages(c("ggplot2", "gplots", "reshape", "scales", "jpg", "png"))

function definiton:

bool br_plot(int num_files, const char *files[], const char *destination, bool show = false)

parameters:

Parameter	Type	Description
num_files	int	Number of .csv files.
files[]	const char *	.csv files created using br_eval.
destination	const char *	Basename for the resulting figures.
show	bool	Open destination.pdf using the system's default PDF viewer. Default is false.

output: (bool) Returns true on success. Returns false on a failure to compile the figures due to a missing, out of date, or incomplete R installation.
see: br_eval

br_plot_detection

Renders detection performance figures for a set of .csv files created by br_eval_detection.

In order of their output, the figures are: 1. Discrete Receiver Operating Characteristic (DiscreteROC) 2. Continuous Receiver Operating Characteristic (ContinuousROC) 3. Discrete Precision Recall (DiscretePR) 4. Continuous Precision Recall (ContinuousPR) 5. Bounding Box Overlap Histogram (Overlap) 6. Average Overlap Table (AverageOverlap) 7. Average Overlap Heatmap (AverageOverlap)

Detection accuracy is measured with overlap fraction = bounding box intersection / union. When computing discrete curves, an overlap >= 0.5 is considered a true positive, otherwise it is considered a false negative. When computing continuous curves, true positives and false negatives are measured fractionally as overlap and 1-overlap respectively.

This function requires a current R installation with the following packages:

install.packages(c("ggplot2", "gplots", "reshape", "scales", "jpg", "png"))

function definition:

bool br_plot_detection(int num_files, const char *files[], const char *destination, bool show = false)

parameters:

Parameter	Type	Description
num_files	int	Number of .csv files.
files[]	const char *	.csv files created using br_eval_detection.
destination	const char *	Basename for the resulting figures.
show	bool	Open destination.pdf using the system's default PDF viewer. Default is false.

output: (bool) Returns true on success. Returns false on a failure to compile the figures due to a missing, out of date, or incomplete R installation.
see: br_eval_detection, br_plot

br_plot_landmarking

Renders landmarking performance figures for a set of .csv files created by br_eval_landmarking.

In order of their output, the figures are: 1. Cumulative landmarks less than normalized error (CD) 2. Normalized error box and whisker plots (Box) 3. Normalized error violin plots (Violin)

Landmarking error is normalized against the distance between two predifined points, usually inter-ocular distance (IOD).

function definition:

bool br_plot_landmarking(int num_files, const char *files[], const char *destination, bool show = false)

parameters:

Parameter	Type	Description
num_files	int	Number of .csv files.
files[]	const char *	.csv files created using br_eval_landmarking.
destination	const char *	Basename for the resulting figures.
show	bool	Open destination.pdf using the system's default PDF viewer. Default is false.

output: (bool) Returns true on success. Returns false on a failure to compile the figures due to a missing, out of date, or incomplete R installation.
see: br_eval_landmarking, br_plot

br_plot_metadata

Renders metadata figures for a set of .csv files with specified columns.

function definition:

bool br_plot_metadata(int num_files, const char *files[], const char *columns, bool show = false)

parameters:

Parameter	Type	Description
num_files	int	Number of .csv files.
files[]	const char *	.csv files created by enrolling templates to .csv metadata files.
columns	const char *	';' seperated list of columns to plot.
show	bool	Open PlotMetadata.pdf using the system's default PDF viewer.

output: (bool) Returns true on success. Returns false on a failure to compile the figures due to a missing, out of date, or incomplete R installation.
see: br_plot

br_progress

Returns current progress from Context::progress.

function definition:
```
float br_progress()
```
parameters: None
output: (float) Returns the completion percentage of the running process
see: br_most_recent_message, br_time_remaining

br_read_pipe

Read and parse arguments from a named pipe. Used by the command line api to implement -daemon, generally not useful otherwise. Guaranteed to return at least one argument. For information on managed returned values see here

function defintion:

void br_read_pipe(const char *pipe, int *argc, char ***argv)

parameters:

Parameter Type Description

pipe const char * Pipe name

argc int * Argument count

argv char *** Argument list
output: (void)

br_scratch_path

Fills the buffer with the value of Context::scratchPath. For information on input string buffers see here.

function defintion:

int br_scratch_path(char * buffer, int buffer_length)

parameters:

Parameter Type Description

buffer char * Buffer for scratch path

buffer_length int Length of buffer.
output: (int) Returns the required size of the input buffer for the most recent message to fit completely
see: br_version

br_sdk_path

Returns the full path to the root of the SDK.

function definition:
```
const char *br_sdk_path()
```
parameters: None
output: (const char *) Returns the full path to the root of the SDK
see: br_initialize

br_get_header

Retrieve the target and query inputs in the BEE matrix header. For information on managed return values see here.

function definition:

void br_get_header(const char *matrix, const char **target_gallery, const char **query_gallery)

parameters:

Parameter Type Description

matrix const char * The BEE matrix file to modify

target_gallery const char ** The matrix target

query_gallery const char ** The matrix query
output: (void)
set: br_set_header

br_set_header

Update the target and query inputs in the BEE matrix header.

function definition:

void br_set_header(const char *matrix, const char *target_gallery, const char *query_gallery)

parameters:

Parameter Type Description

matrix const char * The BEE matrix file to modify

target_gallery const char ** The matrix target

query_gallery const char ** The matrix query
output: (void)
see: br_get_header

br_set_property

Appends a provided value to the global metadata using a provided key.

function definition:

void br_set_property(const char *key, const char *value)

parameters:

Parameter Type Description

key const char * Key to append

value const char * Value to append
output: (void)

br_time_remaining

Returns estimate of time remaining in the current process.

function definition:
```
int br_time_remaining()
```
parameters: None
output: (int) Returns an estimate of the time remaining
see: br_most_recent_message, br_progress

br_train

Trains a provided model's Transform and Distance on the provided input.

function definiton:

void br_train(const char *input, const char *model = "")

parameters:

Parameter	Type	Description
input	const char *	The Gallery to train on.
model	const char *	(Optional) String specifying the binary file to serialize training results to. The trained algorithm can be recovered by using this file as the algorithm. By default the trained algorithm will not be serialized to disk.

output: (void)
see: br_train_n

br_train_n

Convenience function for training on multiple inputs

function definition:

void br_train_n(int num_inputs, const char *inputs[], const char *model = "")

parameters:

Parameter	Type	Description
num_inputs	int	Size of inputs
inputs[]	const char *	An array of galleries to train on.
model	const char *	(Optional) String specifying the binary file to serialize training results to. The trained algorithm can be recovered by using this file as the algorithm. By default the trained algorithm will not be serialized to disk.

output: (void)
see: br_train

br_version

Get the current OpenBR version.

function definition:
```
const char *br_version()
```
parameters: None
output: (const char *) Returns the current OpenBR version
see: br_about, br_scratch_path

br_slave_process

For internal use via ProcessWrapperTransform

function definition:

void br_slave_process(const char * baseKey)

parameters:

Parameter Type Description

baseKey const char * base key for the slave process
output: (void)

br_load_img

Load an image from a string buffer. This is an easy way to pass an image in memory from another programming language to openbr.

function definition:

br_template br_load_img(const char *data, int len)

parameters:

Parameter Type Description

data const char * The image buffer.

len int The length of the buffer.
output: (br_template Returns a br_template loaded with the provided image
see: br_unload_img

br_unload_img

Unload an image to a string buffer. This is an easy way to pass an image from openbr to another programming language.

function definition:

unsigned char* br_unload_img(br_template tmpl)

parameters:

Parameter Type Description

tmpl br_template Pointer to a Template
output: (unsigned char*) Returns a buffer loaded with the image data from tmpl
see: br_load_img

br_template_list_from_buffer

Deserialize a TemplateList from a buffer. Can be the buffer for a .gal file, since they are just a TemplateList serialized to disk.

function definition:

br_template_list br_template_list_from_buffer(const char *buf, int len)

return type: br_template_list
parameters:

Parameter	Type	Description
buf	const char *	The buffer.
len	int	The length of the buffer.

output: (br_template_list) Returns a pointer to a TemplateList created from the buffer.

br_free_template

Free a Template's memory.

function definition:

void br_free_template(br_template tmpl)

parameters:

Parameter Type Description

tmpl br_template Pointer to the Template to free.
output: (void)

br_free_template_list

Free a TemplateList's memory.

function definition:

void br_free_template_list(br_template_list tl)

parameters:

Parameter Type Description

tl br_template_list Pointer to the TemplateList to free.
output: (void)

br_free_output

Free a Output's memory.

function definition:

void br_free_output(br_matrix_output output)

parameters:

Parameter Type Description

output br_matrix_output Pointer to theOutput to free.
output: (void)

br_img_rows

Returns the number of rows in an image.

function definition:
```
int br_img_rows(br_template tmpl)
```
parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.
output: (int) Returns the number of rows in an image

br_img_cols

Returns the number of cols in an image.

function definition:
```
int br_img_cols(br_template tmpl)
```
parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.
output: (int) Returns the number of columns in an image

br_img_channels

Returns the number of channels in an image.

function definition:
```
int br_img_channels(br_template tmpl)
```
parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.
output: (int) Returns the number of channels in an image

br_img_is_empty

Checks if the image is empty.

function definition:
```
bool br_img_is_empty(br_template tmpl)
```
parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.
output: (bool) Returns true if the image is empty, false otherwise.

br_get_filename

Get the name of the file of a provided Template. For information on input string buffers please see here

function definition:

int br_get_filename(char * buffer, int buffer_length, br_template tmpl)

parameters:

Parameter Type Description

buffer char * Buffer to hold the filename

buffer_length int Length of the buffer

tmpl br_template Pointer to a Template.
output: (int) Returns the size of the buffer required to hold the entire file name.

br_set_filename

Set the name of the file for a provided Template.

function definition:

void br_set_filename(br_template tmpl, const char *filename)

parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.

filename const char * New filename for the template.
output: (void)

br_get_metadata_string

Get the metadata value as a string for a provided key in a provided Template.

function definition:

int br_get_metadata_string(char * buffer, int buffer_length, br_template tmpl, const char *key)

parameters:

Parameter	Type	Description
buffer	char *	Buffer to hold the metadata string.
buffer_length	int	length of the buffer.
tmpl	br_template	Pointer to a Template.
key	const char *	Key for the metadata lookup

output: (int) Returns the size of the buffer required to hold the entire metadata string

br_enroll_template

Enroll a Template from the C API!

function definition:

br_template_list br_enroll_template(br_template tmpl)

parameters:

Parameter Type Description

tmpl br_template Pointer to a Template.
output: (br_template_list) Returns a pointer to a TemplateList

br_enroll_template_list

Enroll a TemplateList from the C API!

function definition:

void br_enroll_template_list(br_template_list tl)

parameters:

Parameter Type Description

tl br_template_list Pointer to a TemplateList
output: (void)

br_compare_template_lists

Compare TemplateLists from the C API!

function definition:

br_matrix_output br_compare_template_lists(br_template_list target, br_template_list query)

parameters:

Parameter Type Description

target br_template_list Pointer to a TemplateList

query br_template_list Pointer to a TemplateList
output: (br_matrix_output) Returns a pointer to a MatrixOutput

br_get_matrix_output_at

Get a value in a provided MatrixOutput.

function definition:

float br_get_matrix_output_at(br_matrix_output output, int row, int col)

parameters:

Parameter Type Description

output br_matrix_output Pointer to MatrixOutput

row int Row index for lookup

col int Column index for lookup
output: (float) Returns the value of the MatrixOutput at the provided indexes.

br_get_template

Get a Template from a TemplateList at a specified index.

function definition:

br_template br_get_template(br_template_list tl, int index)

parameters:

Parameter Type Description

tl br_template_list Pointer to a TemplateList

index int Index into the template list. Should be in the range [0,len(tl) - 1].
output: (br_template) Returns a pointer to a Template

br_num_templates

Get the number of Templates in a TemplateList.

function definition:

int br_num_templates(br_template_list tl)

parameters:

Parameter Type Description

tl br_template_list Pointer to a TemplateList
output: (int) Returns the size of the provided TemplateList

br_make_gallery

Initialize a Gallery from a file.

function definition:

br_gallery br_make_gallery(const char *gallery)

parameters:

Parameter Type Description

gallery const char * String location of gallery on disk.
output: (br_gallery) Returns a pointer to a Gallery that has been created from the provided file

br_load_from_gallery

Read a TemplateList from a Gallery.

function definition:

br_template_list br_load_from_gallery(br_gallery gallery)

parameters:

Parameter Type Description

gallery br_gallery Pointer to a Gallery
output: (br_template_list) Returns a pointer to a TemplateList containing the data from the provided Gallery

br_add_template_to_gallery

Write a Template to a Gallery

function definition:

void br_add_template_to_gallery(br_gallery gallery, br_template tmpl)

parameters:

Parameter Type Description

gallery br_gallery Pointer to a Gallery

tmpl br_template Pointer to a Template
output: (void)

br_add_template_list_to_gallery

Write a TemplateList to the Gallery on disk.

function definition:

void br_add_template_list_to_gallery(br_gallery gallery, br_template_list tl)

parameters:

Parameter Type Description

gallery br_gallery Pointer to a Gallery

tl br_template_list Pointer to a TemplateList
output: (void)

br_close_gallery

Close a provided Gallery.

function definition:

void br_close_gallery(br_gallery gallery)

parameters:

Parameter Type Description

gallery br_gallery Pointer to a Gallery
output: (void)

Zhu et al. A Rank-Order Distance based Clustering Algorithm for Face Tagging, CVPR 2011 ↩