Nvvideoconvert buffer transform failed on jetson

When processing png images, nvvideoconvert crashes with the error “buffer transform failed”. If you set compute-hw=1 as recommended in this topic, we get images with artifacts.

**• Hardware Platform: ** Jetson
• DeepStream Version: 6.2
**• JetPack Version: **R35 REVISION: 1.0
**• TensorRT Version: **
**• Issue Type: ** bugs
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)

Download images
png_images.zip (1.9 MB)

Run docker
docker run -it --rm -v pwd:/data nvcr.io/nvidia/deepstream-l4t:6.2-triton

In docker
gst-launch-1.0 filesrc location=/data/foto.png ! pngparse ! pngdec ! queue ! nvvideoconvert ! "video/x-raw(memory:NVMM), format=RGBA" ! nvvideoconvert ! pngenc ! filesink location=/data/foto_res.png

After this command nvvideoconvert crashes with the error:

0:00:00.174602041    92 0xaaaae17e3aa0 ERROR         nvvideoconvert gstnvvideoconvert.c:3768:gst_nvvideoconvert_transform: buffer transform failed
ERROR: from element /GstPipeline:pipeline0/GstQueue:queue0: Internal data stream error.

If we add compute-hw=1
gst-launch-1.0 filesrc location=/data/foto.png ! pngparse ! pngdec ! queue ! nvvideoconvert compute-hw=1 ! "video/x-raw(memory:NVMM), format=RGBA" ! nvvideoconvert ! pngenc ! filesink location=/data/foto_res.png

After this command there is no error after this command, but images with artifacts

  1. The format of your image is RGB that the default mode of Jetson does not support. So you need to set the compute-hw=1.
  2. What do you mean images with artifacts?
  1. How did you determine that the image was GRB? The identify -verbose foto.png command identifies as png with Colorspace sRGB
Command output

Image: foto.png
Format: PNG (Portable Network Graphics)
Mime type: image/png
Class: DirectClass
Geometry: 640x425+0+0
Resolution: 28.34x28.34
Print size: 22.5829x14.9965
Units: PixelsPerCentimeter
Colorspace: sRGB
Type: TrueColor
Base type: Undefined
Endianess: Undefined
Depth: 8-bit
Channel depth:
red: 8-bit
green: 8-bit
blue: 8-bit
Channel statistics:
Pixels: 272000
min: 0 (0)
max: 255 (1)
mean: 198.25 (0.77745)
standard deviation: 51.0015 (0.200006)
kurtosis: 6.06593
skewness: -2.6742
entropy: 0.764968
min: 0 (0)
max: 255 (1)
mean: 194.701 (0.763533)
standard deviation: 59.0813 (0.231691)
kurtosis: 4.38842
skewness: -2.42687
entropy: 0.754563
min: 0 (0)
max: 255 (1)
mean: 199.013 (0.780445)
standard deviation: 57.748 (0.226463)
kurtosis: 4.53538
skewness: -2.4602
entropy: 0.736281
Image statistics:
min: 0 (0)
max: 255 (1)
mean: 197.321 (0.77381)
standard deviation: 55.9436 (0.219387)
kurtosis: 4.95311
skewness: -2.51715
entropy: 0.751937
Rendering intent: Perceptual
Gamma: 0.454545
red primary: (0.64,0.33)
green primary: (0.3,0.6)
blue primary: (0.15,0.06)
white point: (0.3127,0.329)
Background color: white
Border color: srgb(223,223,223)
Matte color: grey74
Transparent color: black
Interlace: None
Intensity: Undefined
Compose: Over
Page geometry: 640x425+0+0
Dispose: Undefined
Iterations: 0
Compression: Zip
Orientation: Undefined
comment: AppleMark

date:create: 2023-07-31T11:10:20+07:00
date:modify: 2023-07-28T15:53:46+07:00
icc:copyright: Copyright 2000 Adobe Systems Incorporated
icc:description: Adobe RGB (1998)
png:bKGD: chunk was found (see Background color, above)
png:cHRM: chunk was found (see Chromaticity, above)
png:iCCP: chunk was found
png:IHDR.bit-depth-orig: 8
png:IHDR.bit_depth: 8
png:IHDR.color-type-orig: 2
png:IHDR.color_type: 2 (Truecolor)
png:IHDR.interlace_method: 0 (Not interlaced)
png:IHDR.width,height: 640, 425
png:pHYs: x_res=2834, y_res=2834, units=1
png:text: 5 tEXt/zTXt/iTXt chunks were found
signature: 1433c58c81eabef72c45eb03229268a3499551feefe591b03ccba900cd44c59b

Profile-icc: 560 bytes
filename: foto.png
verbose: true
Tainted: False
Filesize: 378557B
Number pixels: 272000
Pixels per second: 9.06667MB
User time: 0.020u
Elapsed time: 0:01.030
Version: ImageMagick 6.9.10-23 Q16 aarch64 20190101 https://imagemagick.org

  1. Image after executing the command
    gst-launch-1.0 filesrc location=/data/foto.png ! pngparse ! pngdec ! queue ! nvvideoconvert compute-hw=1 ! "video/x-raw(memory:NVMM), format=RGBA" ! nvvideoconvert ! pngenc ! filesink location=/data/foto_res.png

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

I have tried on my Orin, it works well like below with the command you attached.
What’s the specific model of your platform? And some of your software versions may have compatibility issues with DeepStream 6.2. You can refer to the link below: Jetson model Platform and OS Compatibility.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.