Browse 30,046+ expert-crafted prompts
Showing 20 prompts
Build a robust barcode and QR code reader that handles angles, lighting, blur, and multiple codes per image.
## CONTEXT A developer needs to read barcodes and QR codes from images or a live camera, reliably under poor lighting, odd angles, and partial occlusi...
Analyze document images to detect structure: text blocks, tables, figures, and reading order for downstream extraction.
## CONTEXT A developer processes complex documents (forms, invoices, reports) and needs to understand their layout before extracting content. Plain OC...
Design an efficient image annotation workflow with guidelines, tooling, quality control, and model-assisted labeling.
## CONTEXT A team must label thousands of images for a vision project. They want a workflow that produces consistent, high-quality annotations efficie...
Stress-test a vision model against corruptions, distribution shift, and adversarial inputs before deployment.
## CONTEXT A vision model performs well on the test set but its real-world robustness is unknown. Before deployment, the developer needs to test it ag...
Compute optical flow and analyze motion in video for stabilization, motion detection, and velocity estimation.
## CONTEXT A developer needs to analyze motion between video frames for tasks like stabilization, activity detection, or velocity estimation. They mus...
Build a human or object pose estimation pipeline with keypoint detection, smoothing, and action understanding.
## CONTEXT A developer needs to estimate body or object keypoints from images or video, for fitness, sports, animation, or ergonomics. They need accur...
Build a reverse-image and similarity search system using embeddings, a vector index, and relevance evaluation.
## CONTEXT A developer needs to find visually similar images at scale, for duplicate detection, product search, or content moderation. They need to em...
Assess and filter image quality automatically to gate inputs into a vision pipeline by sharpness, exposure, and noise.
## CONTEXT A vision pipeline receives images of inconsistent quality. Blurry, over/under-exposed, or noisy inputs degrade results. The developer wants...
Build a classifier that works with only a handful of examples per class using metric learning and foundation embeddings.
## CONTEXT A developer has only a few labeled examples per class but needs a working classifier. Standard supervised training overfits, so they need f...
Build a background removal and alpha matting pipeline with clean edges, hair detail, and compositing.
## CONTEXT A developer needs to remove backgrounds from images for e-commerce, profile photos, or compositing. Hard segmentation leaves jagged edges, ...
Upscale and restore degraded images with super-resolution, denoising, and artifact removal while avoiding hallucination.
## CONTEXT A developer needs to enhance low-resolution, noisy, or compressed images for display or downstream processing. They must improve perceived ...
Profile and optimize an end-to-end vision pipeline to find and fix bottlenecks in I/O, preprocessing, and inference.
## CONTEXT A vision pipeline is too slow but the developer does not know where time goes. The bottleneck could be image loading, preprocessing, infere...
Explain and interpret vision model predictions with saliency maps, concept attribution, and counterfactual analysis.
## CONTEXT A vision model makes predictions that stakeholders do not trust or understand. The developer needs to explain what the model looks at, whet...
Calibrate a camera and apply geometric transforms for undistortion, measurement, and multi-view alignment.
## CONTEXT A developer needs accurate geometry from camera images: removing lens distortion, measuring real-world dimensions, or aligning multiple vie...
Plan and generate synthetic image data to fill gaps, balance classes, and reduce real-data labeling cost.
## CONTEXT A team lacks enough real labeled images for rare cases or struggles with labeling cost. They want synthetic data from rendering, GANs/diffu...
Integrate a vision-language model for captioning, VQA, or grounding with proper prompting and evaluation.
## CONTEXT A developer wants to use a vision-language model (CLIP, BLIP, LLaVA, or a hosted VLM) for tasks like captioning, visual question answering,...
Find whether a Linux performance problem is CPU, memory, disk, or network bound.
## CONTEXT You are investigating a Linux server that is slow or unresponsive under load. The hardest part is determining which resource is the actual ...
Choose between cron and systemd timers and configure reliable scheduled jobs.
## CONTEXT You are scheduling recurring tasks on a Linux server and must decide between cron and systemd timers, then configure the job so it runs rel...
Diagnose stuck, zombie, or runaway processes and apply the right signal safely.
## CONTEXT You are dealing with a misbehaving process on Linux: a daemon that will not stop, a zombie cluttering the process table, a runaway consumin...
Harden an OpenSSH server configuration for secure, key-based, least-exposure access.
## CONTEXT You are securing the SSH service on a Linux server, the most commonly attacked entry point on internet-facing hosts. A default configuratio...
Scroll for more