MAMuhammad Ariel Shakaramiroinshaka-ai.hashnode.dev·2d ago · 8 min readI Shrank an Object Detection Model 4x — Its Accuracy Went Up Instead. Here's What HappenedIf you've ever tried deploying a computer vision model on an ordinary CPU (no fancy GPU server), you've probably hit the same wall: the model runs, but it's slow. Too slow for anything real-time. I re00
MAMuhammad Ariel Shakaramiroinshaka-ai.hashnode.dev·2d ago · 7 min readSaya Kompres Model Deteksi Objek 4x Lebih Kecil, Akurasinya Malah Naik. Ini CeritanyaKalau kamu pernah coba deploy model computer vision ke perangkat yang CPU-nya biasa saja (bukan server dengan GPU mahal), kamu pasti pernah mentok di masalah yang sama: model-nya jalan, tapi lambat. T00
MAMuhammad Ariel Shakaramiroinshaka-ai.hashnode.dev·4d ago · 7 min readInstance Segmentation: Counting Cars from Public CCTV with YOLOv8 segTable of Contents Why Isn't a Box Enough? Guess First Semantic vs Object Detection vs Instance Segmentation Why Instance Segmentation Matters From Mask R-CNN to YOLO-Segmentation Building the Mo04A
MAMuhammad Ariel Shakaramiroinshaka-ai.hashnode.dev·4d ago · 6 min readInstance Segmentation: Menghitung Mobil dari CCTV Publik dengan YOLOv8 segDaftar Isi Kenapa Bukan Cuma Kotak? Coba Tebak Dulu Semantic vs Object Detection vs Instance Segmentation Kenapa Instance Segmentation Dibutuhkan Dari Mask R-CNN ke YOLO-Segmentation Membangun M00
MAMuhammad Ariel Shakaramiroinshaka-ai.hashnode.dev·5d ago · 7 min readFace Recognition: Turning a Classifier Into an Embedding That Can Learn New FacesPhones that unlock just by looking at you, office systems that recognize whoever just walked through the door: the trick behind both is rarely "just train a classifier." This post walks through why, a00
