acldvppJpegDecodeAsync

Applicability

Product	Supported
Atlas 350 Accelerator Card	☓
Atlas A3 training product / Atlas A3 inference product	√
Atlas A2 training product / Atlas A2 inference product	√
Atlas 200I/500 A2 inference product	√
Atlas inference product	√
Atlas training product	√

Function Description

Decodes a .jpg, .jpeg, .JPG, or .JPEG image. This API is asynchronous.

Prototype

      
           aclError acldvppJpegDecodeAsync(acldvppChannelDesc *channelDesc,
const void *data,
uint32_t size,
acldvppPicDesc *outputDesc,
aclrtStream stream)

Parameters

Parameter	Input/Output	Description
channelDesc	Input	Pointer to the channel description. The value must be the same as that of channelDesc specified when the acldvppCreateChannel API is called to create a channel.
data	Input	Pointer to the input buffer address.
size	Input	Actual size of the input image, in bytes.
outputDesc	Input/Output	Pointer to the output image description. When outputDesc is used as an input, the user needs to: Call the acldvppCreatePicDesc API to create the image description. Call acldvppSetPicDesc series to set the buffer address, buffer size, image format, widthStride, and heightStride of the output image. Call acldvppJpegPredictDecSize to predict the output buffer size in advance. Call acldvppJpegGetImageInfo or acldvppJpegGetImageInfoV2 to obtain the width and height of the input JPEG image. For details about the output format and width and height alignment requirements, see Functions and Restrictions. When outputDesc is used as an output, you can obtain the decoding output from the buffer address and call the acldvppGetPicDesc series to obtain the width and height of the output image.
stream	Input	Stream. For details about the type definition, see aclrtStream.

Returns

If the return value is 0, the task is successfully delivered. If the return value is not 0, the task fails to be delivered. For details, see aclError.

Restrictions

If the decoded output image data needs to be used in subsequent operations (for example, cropping and resizing with VPC), you are advised to call acldvppGetPicDesc series APIs to obtain the width and height of the output image after decoding, because the width and height may be aligned during decoding. For example, if a JPEG (444) source image is decoded into a YUV420SP output image and the width and height of the JPEG (444) source image are odd numbers, theoretically, the width and height of the decoded YUV420SP output image should be odd numbers. However, the YUV420SP format requires that the width and height of the image be even numbers. In this case, the JPEGD rounds down the odd width and height to the nearest multiple of 2.

This restriction applies to the following product models:

Atlas inference product

Atlas 200I/500 A2 inference product

Atlas A2 training product / Atlas A2 inference product

Atlas A3 training product / Atlas A3 inference product