Fast J2K Viewer on GPU for geospatial applications

GPU JPEG2000 codec from Fastvideo is based on NVIDIA technology. This is full, performance-oriented implementation of JPEG2000. We got fast JPEG2000 compression and decompression on the GPU due to parallel implementation and thorough optimization of JPEG2000 algorithm. Our JPEG2000 encoder on GPU is the fastest on the market.

fast j2k viewer

Key Features of Fast J2K Viewer on CUDA

  • Input formats: JP2, J2K, JPX
  • GMLJP2 compliance with OGC
  • Geographic coordinates computation, associated to a pixel (column, row)
  • Partial support of JPEG 2000 part 2
  • J2K decoding for grayscale and color images with arbitrary width and height
  • Lossy (wavelet CDF 9/7) and lossless (wavelet CDF 5/3) image decoding
  • Bit depth: 8-16 bits per channel
  • Number of decomposition levels: 1–12
  • Code-block size 16×16, 32×32 or 64×64, etc.
  • Chroma subsampling modes: 4:4:4, 4:2:2, 4:2:0
  • Tile support
  • Window mode for J2K decoder
  • Data input: images from HDD/RAID/SSD
  • Data output: final uncompressed images on the screen
  • Optimized for the latest NVIDIA GPUs
  • Time measurements for jp2/jpx decoding
  • Compatible with 64-bit Windows-7/10, Linux Ubuntu, L4T

Test images for J2K decoding

These are images which we have utilized for J2K decoding:

At the following link you can download Fast J2K Viewer

Roadmap for Fast J2K Viewer

  • JPX format with GMLJP2 markup (JPEG 2000 Part 2) - done
  • Geographic coordinates computation, associated to a pixel (column, row) - done
  • GDAL support - done
  • Performance optimization for Fast J2K Viewer on GPU - in progress

Contact Form

This form collects your name and email. Check out our Privacy Policy on how we protect and manage your personal data.