📰 Dev.to · Marco Rinaldi
8 articles · Updated every 3 hours · View all reads
All
Articles 96,459Blog Posts 112,951Tech Tutorials 24,348Research Papers 20,269News 15,465
⚡ AI Lessons
Dev.to · Marco Rinaldi
3w ago
Our event-camera detector lost 6 mAP to a badly chosen accumulation window
TL;DR: We spent three weeks chasing a 6 mAP regression in an event-camera object detector. The model...
Dev.to · Marco Rinaldi
3w ago
VLM-scored calibration sets for INT8 quantisation, routed through Bifrost
TL;DR: We pick the 512 hardest images for INT8 PTQ calibration by scoring a candidate pool with a...
Dev.to · Marco Rinaldi
3w ago
QAT vs PTQ on our edge vision model: 6 months of A/B data
TL;DR: We ran post-training quantisation (PTQ) and quantisation-aware training (QAT) side by side on...
Dev.to · Marco Rinaldi
4w ago
Capping VLM spend per CV researcher: hierarchical budgets in practice
TL;DR: Our 11-person CV team at Prophesee was burning through €3-4k weeks of VLM spend on dataset...
Dev.to · Marco Rinaldi
🧠 Large Language Models
⚡ AI Lesson
1mo ago
Auto-labelling 1.2M robotics frames with VLMs: a failover story
TL;DR: We needed to caption 1.2M reconstructed event-camera frames using vision-language models for...
Dev.to · Marco Rinaldi
📐 ML Fundamentals
⚡ AI Lesson
1mo ago
Quantising event-camera networks to run under 1MB on a Cortex-M7
TL;DR: I shrunk a gesture-recognition model for a Prophesee EVK4 event camera from 4.2MB down to...
Dev.to · Marco Rinaldi
1mo ago
Routing Event-Camera Pipelines Through an LLM Gateway: A Field Report
TL;DR: We added a vision-language stage to an event-camera pipeline at Prophesee and the LLM provider...
Dev.to · Marco Rinaldi
2mo ago
Event cameras and edge inference: why the frame-based mindset is still holding us back
So, the thing is, most edge inference pipelines for computer vision are built around a mental model...
DeepCamp AI