nvJPEG is a high-performance GPU-accelerated library for JPEG decoding. nvJPEG supports decoding of single and batched images, color space conversion, multiple phase decoding, and hybrid decoding using both CPU and GPU. Applications that rely on nvJPEG for decoding deliver higher throughput and lower latency JPEG decode compared CPU-only decoding.
nvJPEG provides low-latency decoder for common JPEG formats used in computer vision applications such as image classification, object detection and image segmentation. For deep learning training applications, nvJPEG can accelerate data loading and pre-processing with GPU-accelerated augmentation such as translation, zoom, scale, corp, flip and others. For applications that demand low-latency deep learning inference, nvJPEG can be used to perform JPEG decoding and resizing in real-time.
nvJPEG is available as part of all CUDA Toolkit versions, starting with CUDA 10: Download latest CUDA Toolkit
If you are using CUDA Toolkit 9.0, please download the: nvJPEG pre-release
- nvJPEG user guide (Documentation)
- NVIDIA Data Loading Library for fast data loading and augmentation
- Hybrid decoding using both the CPU and the GPU
- Single image and batched image decoding
- Color space conversion to RGB, BGR, RGBI, BGRI, and YUV
- Single and multi phase decoding