Skeleton interface (base C API)

Sends notifications that skeleton reports (aka tracker reports) are complete. More...

Typedefs
typedef struct OSVR_SkeletonDeviceInterfaceObject *	OSVR_SkeletonDeviceInterface
	Opaque type used in conjunction with a device token to send data on Skeleton Interface.

Functions
OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode	osvrDeviceSkeletonConfigure (OSVR_INOUT_PTR OSVR_DeviceInitOptions opts, OSVR_OUT_PTR OSVR_SkeletonDeviceInterface *iface, OSVR_IN_READS(len) const char *jsonDescriptor) OSVR_FUNC_NONNULL((1
	Specify that your device will implement the Skeleton interface. More...
OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode	osvrDeviceSkeletonComplete (OSVR_IN_PTR OSVR_SkeletonDeviceInterface iface, OSVR_IN OSVR_ChannelCount sensor, OSVR_IN_PTR OSVR_TimeValue const *timestamp) OSVR_FUNC_NONNULL((1
	Report that transmission of tracker reports for given skeleton sensor is complete. More...
OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode	osvrDeviceSkeletonUpdateSpec (OSVR_IN_PTR OSVR_SkeletonDeviceInterface iface, OSVR_IN_READS(len) const char *spec) OSVR_FUNC_NONNULL((1
	If device detects another skeleton and/or change in existing articulation specification, then it needs to update the spec with the client as well. More...

Detailed Description

Sends notifications that skeleton reports (aka tracker reports) are complete.

Skeleton interface is implemented as a device that exposes a tracker interface for each skeleton element. For each element, a separate tracker report will be sent with an updated pose followed by skeletonComplete. and the client can request a complete report of entire skeleton (all reported elements), as well as subsets. Note that since it exposes a device via multiple interfaces, you will need to "configure" both interfaces (Tracker and Skeleton).

Function Documentation

osvrDeviceSkeletonComplete()

OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode osvrDeviceSkeletonComplete	(	OSVR_IN_PTR OSVR_SkeletonDeviceInterface	iface,
		OSVR_IN OSVR_ChannelCount	sensor,
		OSVR_IN_PTR OSVR_TimeValue const *	timestamp
	)

Report that transmission of tracker reports for given skeleton sensor is complete.

This method should be called after device reports updated poses for skeleton elements for given skeleton sensor. For example, if device is reporting two hands (two skeleton sensors), it should report all poses for left hand followed by osvrDeviceSkeletonComplete(leftSensor) call and then report all poses for right hand followed by osvrDeviceSkeletonComplete(rightSensor). This ensures that client receives consistent, single-frame reports and avoids bone stretching.

Parameters

dev	Device token
iface	Skeleton Interface
sensor	Sensor number
timestamp	The same timestamp as for your tracker reports

osvrDeviceSkeletonConfigure()

OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode osvrDeviceSkeletonConfigure	(	OSVR_INOUT_PTR OSVR_DeviceInitOptions	opts,
		OSVR_OUT_PTR OSVR_SkeletonDeviceInterface *	iface,
		OSVR_IN_READS(len) const char *	jsonDescriptor
	)

Specify that your device will implement the Skeleton interface.

Parameters

	opts	The device init options object.
[out]	iface	An interface object you should retain with the same lifetime as the device token in order to send messages conforming to a Skeleton interface.
	jsonDescriptor	A device descriptor json that contains skeleton spec Note that the number of skeleton sensors is not the same as number of skeleton elements. For example if plugin device reports two hands that will be 2 separate skeleton sensors because hands are not connected via common parent. If device can report an entire skeleton then it should only report one sensor.

osvrDeviceSkeletonUpdateSpec()

OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode OSVR_PLUGINKIT_EXPORT OSVR_ReturnCode osvrDeviceSkeletonUpdateSpec	(	OSVR_IN_PTR OSVR_SkeletonDeviceInterface	iface,
		OSVR_IN_READS(len) const char *	spec
	)

If device detects another skeleton and/or change in existing articulation specification, then it needs to update the spec with the client as well.

During the skeleton interface initialization this is performed once using the spec provided in the device descriptor. This should be used for unbounded skeletons where the articulation spec may change over time. Devices with bounded skeletons don't need to use this method because the articulation spec remains the same throughout plugin runtime

Parameters

dev	Device token
iface	Skeleton Interface
spec	Updated/New Skeleton Articulation Spec in JSON format
timestamp	The same timestamp as for your tracker reports