// the find
amusi/awesome-object-detection
Awesome Object Detection based on handong1587 github: https://handong1587.github.io/deep_learning/2015/10/09/object-detection.html
A link-dump 'awesome list' of object detection papers and their implementations, organized by architecture family (R-CNN through M2Det, plus a toolbox section at the bottom). It's for someone who wants a map of the pre-transformer detection literature, not a working codebase.
Grouping by lineage (R-CNN → Fast R-CNN → Faster R-CNN → FPN → RetinaNet, etc.) makes the historical arc of the field legible in a way a flat paper list wouldn't. Each entry links straight to the arxiv paper and usually several independent implementations across Caffe, TensorFlow, PyTorch, and MXNet, so you can compare framework-specific ports of the same idea. The toolbox section at the end correctly points past the list itself to Detectron2 and mmdetection, the actual actively maintained frameworks people should be using.
Dead since the 2022-12-17 push: essentially nothing from 2023 onward, so DETR-family and transformer-based detectors, and the YOLOv8-v11 generation, are missing or barely mentioned. It's a flat list with zero curation signal — no indication of which of these 100+ linked repos are still maintained versus abandoned grad-student code from 2016, and several of the linked personal repos are almost certainly bit-rotted. Formatting is inconsistent (full-width colons like 'intro：', broken relative links to a camo-proxied image, mixed spacing) suggesting it was pasted in from another source and never cleaned up. There's no working code or example in this repo itself — it's only useful as a set of outbound links, and arxiv/GitHub link rot is a real risk given the age.