⚡ Optimize ImageSegmenter memory allocation#224
Conversation
Pre-allocate buffers for `cv2.resize` and `cv2.cvtColor` operations in `ImageSegmenter` to avoid repetitive memory allocation per frame. This improves performance and reduces memory churn. Measured improvement: ~14% increase in FPS (from ~59.5 FPS to ~68.2 FPS) on benchmark. Co-authored-by: fangfufu <2323403+fangfufu@users.noreply.github.com>
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with New to Jules? Learn more at jules.google/docs. For security, I will only act on instructions from the user who triggered this task. |
|
fangfufu
deleted the
optimize-image-segmenter-memory-17371324163120674587
branch
1 participant
This PR optimizes the
ImageSegmenter.segmentmethod inlfbw/lfbw.pyby pre-allocating buffers for image resizing and color conversion. Previously, these operations allocated new memory for every frame, leading to unnecessary overhead.Key changes:
self.mp_frame_bufferandself.mask_upscaled_bufferinImageSegmenter.__init__.ImageSegmenter.segmentto use these buffers via thedstparameter incv2.resizeandcv2.cvtColor.cv2.cvtColorandcv2.GaussianBluroperate in-place where appropriate.This change is transparent to the caller and maintains the existing functionality while improving performance.
Performance Benchmark:
PR created automatically by Jules for task 17371324163120674587 started by @fangfufu