6、物体分类
算法简介
CSK6 大模型开发套件可通过摄像头抓拍画面并对该画面进行物体分类识别,支持包括苹果、床、键盘等100+种物体的识别。通过触摸屏可进行取景与抓拍控制,完成拍照后将自动识别并在屏幕上显示识别结果。
本示例演示在开发套件上运行基于pytorch-cifar100训练的resnet18物体分类模型,支持通过摄像头拍照并完成对指定物品的分类识别。
本模型支持以下物体的分类识别:
"apple", "aquarium_fish", "baby", "bear", "beaver", "bed",
"bee", "beetle", "bicycle", "bottle", "bowl", "boy",
"bridge", "bus", "butterfly", "camel", "can", "castle",
"caterpillar", "cattle", "chair", "chimpanzee", "clock", "cloud",
"cockroach", "couch", "crab", "crocodile", "cup", "dinosaur",
"dolphin", "elephant", "flatfish", "forest", "fox", "girl",
"hamster", "house", "kangaroo", "keyboard", "lamp", "lawn_mower",
"leopard", "lion", "lizard", "lobster", "man", "maple_tree",
"motorcycle", "mountain", "mouse", "mushroom", "oak_tree", "orange",
"orchid", "otter", "palm_tree", "pear", "pickup_truck", "pine_tree",
"plain", "plate", "poppy", "porcupine", "possum", "rabbit",
"raccoon", "ray", "road", "rocket", "rose", "sea",
"seal", "shark", "shrew", "skunk", "skyscraper", "snail",
"snake", "spider", "squirrel", "streetcar", "sunflower", "sweet_pepper",
"table", "tank", "telephone", "television", "tiger", "tractor",
"train", "trout", "tulip", "turtle", "wardrobe", "whale",
"willow_tree", "wolf", "woman", "worm"
本示例基于开源项目进行移植适配,仅用于 CV 能力的验证与评估,不代表适用于商用项目。
功能展示
- 点击屏幕 翻转 按钮,可实现对摄像头的影像预览翻转,可根据摄像头是否安装在开发板背面进行切换
- 点击屏幕 TAKE 按钮,可对当面画面进行拍照并识别
SDK资源下载
大模型拍照识图 :https://cloud.listenai.com/CSKG962172/duomotai_ap/-/tree/feature/awe_open/apps/LLM_pic
其他功能SDK下载地址:https://cloud.listenai.com/CSKG962172/duomotai_ap/-/tree/master/
- 坐姿检测: apps目录下,工程目录名称为 lcd_spd
- 人脸识别: apps目录下,工程目录名称为 fd
- 活体识别: apps目录下,工程目录名称为 fdh
- 头肩跟随+手势识别: apps目录下,工程目录名称为 hsd
- 物体分类: apps目录下,工程目录名称为 resnet18
已打包好的DEMO固件下载:
补充开发板信息
开发板具备丰富语音图像功能与硬件外设的开发板,采用有着丰富组件生态的 Zephyr RTOS 作为操作系统,默认配套开箱即玩的 AI 应用,也可以配合聆思的模型训练推理工具 LNN 将自己的算法模型部署至芯片上,构建自己的 AI 应用,开发板详情参考:https://docs2.listenai.com/x/nTn9kMMCU