Skip to content

DepthEstimator

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:70

readonly inputSize: number

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:72

Width and height of every returned map.


readonly schemas: object

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:75

Buffer layouts for encodeEstimateDepth, at this runner’s size. Pass one to root.createBuffer to make a buffer of the right type and size.

readonly input: DepthInput

readonly output: DepthMapData

dispose(): void

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:87

Frees the weights and every GPU buffer the runner holds.

void


encodeEstimateDepth(options): void

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:85

Records a depth estimate from GPU memory into GPU memory, with no CPU copy. A texture source is fitted like an ImageBuffer; a buffer source is the model input as-is. A texture target gets the map stretched to its size in every channel; a buffer target gets the map’s values. Work submitted after it sees the result.

EncodeEstimateDepthOptions

void

RunntimeError INVALID_ARGUMENT for a resource of the wrong format, usage or size.


estimateDepth(image): Promise<DepthMap>

Defined in: packages/runntime/src/zoo/tasks/cv/depthEstimation.ts:77

Image in, its depth map out.

ImageBuffer

Promise<DepthMap>