Triton-Augment: GPU Kernel Fusion for 5-73x Faster Image/video Augmentation
Postedabout 2 months agoActiveabout 2 months ago
github.comTechstory
calmpositive
Debate
0/100
GPU OptimizationImage/video ProcessingTriton
Key topics
GPU Optimization
Image/video Processing
Triton
The author shares Triton-Augment, a library for faster image/video augmentation using GPU kernel fusion, and receives a positive initial response.
Snapshot generated from the HN discussion
Discussion Activity
Light discussionFirst comment
N/A
Peak period
2
0-1h
Avg / period
2
Key moments
- 01Story posted
Nov 18, 2025 at 11:28 AM EST
about 2 months ago
Step 01 - 02First comment
Nov 18, 2025 at 11:28 AM EST
0s after posting
Step 02 - 03Peak activity
2 comments in 0-1h
Hottest window of the conversation
Step 03 - 04Latest activity
Nov 18, 2025 at 11:33 AM EST
about 2 months ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
ID: 45968400Type: storyLast synced: 11/18/2025, 4:29:49 PM
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
The core issue is the "Global Memory Tax": Sequential transforms (Crop, Jitter, Normalize) force the GPU to repeatedly read/write intermediate tensors to VRAM. This kills performance.
The Solution: I use Triton to fuse the entire augmentation pipeline into a single, highly-optimized GPU kernel. This eliminates all intermediate memory I/O.
The Results:
Video: Up to 73.7x faster than Kornia on 5D video tensors.
Image: 8.1x average speedup (up to 12x) over Torchvision v2.
It's designed as a drop-in replacement for your existing Compose pipeline. Check out the GitHub repository for the full API and detailed benchmarks.
I'm focused on developing the next phase (Resize, Rotation, etc.) and welcome any feedback on the kernels or usage patterns!
GitHub: https://github.com/yuhezhang-ai/triton-augment