rt-detr-finetuned-cppe-5-3k-steps Open-source Object Detection Model - Accurately Identify Medical Protective Equipment

Rt Detr Finetuned Cppe 5 3k Steps

Developed by qubvel-hf

This model is an object detection model fine-tuned on the medical protective equipment dataset CPPE-5, based on PekingU/rtdetr_r50vd_coco_o365

Object Detection

Transformers

Open Source License:Apache-2.0 #Medical Protective Equipment Detection #Real-time Object Detection #Small Object Recognition Optimization

Downloads 13

Release Time : 12/2/2024

Model Overview

Used for detecting protective equipment in medical scenarios, including five categories: protective clothing, face shields, gloves, goggles, and masks

Model Features

Medical Protective Equipment Detection

Specialized detection capability optimized for medical scenarios

Multi-scale Object Detection

Supports detection of small, medium, and large-sized objects

Efficient Training

Optimized training process using mixed precision training and linear learning rate scheduler

Model Capabilities

Medical protective equipment recognition

Multi-category object detection

Real-time object detection

Use Cases

Medical Safety

Personal Protective Equipment Compliance Check

Automatically detects whether medical staff are wearing complete protective equipment

Protective clothing detection mAP reached 0.4438

Medical Facility Safety Monitoring

Real-time monitoring of protective equipment usage in hospitals and other facilities

Overall recall@100 reached 0.5014

🚀 rt-detr-finetuned-cppe-5-3k-steps

This model is a fine - tuned version of PekingU/rtdetr_r50vd_coco_o365 on the cppe - 5 dataset. It is designed for object - detection tasks in the field of computer vision.

✨ Features

This fine - tuned model can achieve the following results on the evaluation set:

Loss: 9.1012
Map: 0.2813
Map 50: 0.5271
Map 75: 0.2685
Map Small: 0.0879
Map Medium: 0.2399
Map Large: 0.4613
Mar 1: 0.3061
Mar 10: 0.4664
Mar 100: 0.5014
Mar Small: 0.2985
Mar Medium: 0.4465
Mar Large: 0.6698
Map Coverall: 0.4438
Mar 100 Coverall: 0.6815
Map Face Shield: 0.2983
Mar 100 Face Shield: 0.4924
Map Gloves: 0.2305
Mar 100 Gloves: 0.4817
Map Goggles: 0.1591
Mar 100 Goggles: 0.3969
Map Mask: 0.275
Mar 100 Mask: 0.4547

📚 Documentation

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e - 05
train_batch_size: 8
eval_batch_size: 8
seed: 1337
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon = 1e - 08 and optimizer_args = No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 30.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Map	Map 50	Map 75	Map Small	Map Medium	Map Large	Mar 1	Mar 10	Mar 100	Mar Small	Mar Medium	Mar Large	Map Coverall	Mar 100 Coverall	Map Face Shield	Mar 100 Face Shield	Map Gloves	Mar 100 Gloves	Map Goggles	Mar 100 Goggles	Map Mask	Mar 100 Mask
41.0803	1.0	107	14.7670	0.0685	0.1413	0.0589	0.0211	0.0496	0.1045	0.0968	0.2251	0.2568	0.127	0.201	0.3699	0.2045	0.3964	0.0514	0.3013	0.0134	0.1777	0.0031	0.1938	0.0701	0.2147
18.7124	2.0	214	11.5242	0.1664	0.3287	0.148	0.0451	0.1291	0.265	0.1963	0.3451	0.3986	0.2461	0.3493	0.5659	0.3441	0.5757	0.1158	0.4367	0.1193	0.3446	0.0676	0.2754	0.1854	0.3604
16.6913	3.0	321	12.2486	0.1399	0.3041	0.1087	0.029	0.1187	0.2072	0.1663	0.3153	0.3776	0.1937	0.3363	0.5251	0.3001	0.5545	0.1189	0.443	0.0832	0.3326	0.0431	0.2092	0.1543	0.3484
15.7025	4.0	428	11.5488	0.1819	0.3681	0.1619	0.0474	0.1853	0.2567	0.2206	0.351	0.3984	0.1976	0.3882	0.4911	0.3217	0.5752	0.1295	0.3709	0.1665	0.3241	0.1069	0.3292	0.1847	0.3924
14.7758	5.0	535	12.9331	0.1605	0.3126	0.1484	0.0402	0.1512	0.2597	0.1985	0.3135	0.3435	0.1853	0.3037	0.4472	0.2383	0.4365	0.14	0.3304	0.1715	0.3402	0.0775	0.2569	0.1754	0.3538
14.4905	6.0	642	10.1206	0.219	0.4286	0.1979	0.0502	0.1962	0.3665	0.2516	0.4058	0.4494	0.216	0.4008	0.5944	0.404	0.673	0.1247	0.3975	0.2031	0.4402	0.1247	0.2985	0.2387	0.4378
13.4909	7.0	749	9.6988	0.245	0.4741	0.2326	0.0588	0.2417	0.3664	0.2647	0.4201	0.4572	0.2248	0.4316	0.5821	0.4571	0.6806	0.1374	0.3405	0.1979	0.4473	0.159	0.3831	0.2736	0.4347
13.2245	8.0	856	9.7353	0.2375	0.4622	0.227	0.0528	0.2347	0.3748	0.2658	0.4291	0.4815	0.2621	0.4423	0.6434	0.4127	0.6707	0.1807	0.4696	0.2293	0.4821	0.1382	0.3462	0.2265	0.4391
12.7879	9.0	963	9.2718	0.2604	0.4832	0.2502	0.0654	0.2323	0.4055	0.2818	0.4304	0.4788	0.2339	0.4387	0.6353	0.4797	0.6959	0.1558	0.4329	0.2371	0.4884	0.1658	0.3354	0.2633	0.4413
12.2499	10.0	1070	9.5461	0.2547	0.4956	0.2404	0.0582	0.2249	0.418	0.2725	0.443	0.4933	0.2713	0.4359	0.6505	0.4224	0.7059	0.2289	0.4544	0.2279	0.4915	0.1501	0.3862	0.2443	0.4284
12.1284	11.0	1177	9.6199	0.2611	0.5056	0.2322	0.0731	0.2459	0.3958	0.2646	0.4262	0.4704	0.2817	0.4256	0.598	0.4442	0.6568	0.211	0.4354	0.2404	0.475	0.1468	0.3446	0.2633	0.44
11.9831	12.0	1284	9.3471	0.2556	0.5007	0.2397	0.0889	0.2476	0.3971	0.2822	0.4573	0.5045	0.3054	0.4747	0.6499	0.4483	0.6806	0.1954	0.4911	0.2084	0.4786	0.1867	0.4292	0.2391	0.4431
11.8266	13.0	1391	9.5850	0.2329	0.4647	0.2113	0.0657	0.2009	0.3955	0.2667	0.4193	0.4586	0.2653	0.3996	0.6083	0.4132	0.6329	0.1537	0.4063	0.2098	0.4768	0.1427	0.3569	0.2451	0.42
11.6433	14.0	1498	9.7106	0.2353	0.472	0.2148	0.0697	0.2149	0.3916	0.2711	0.4243	0.4627	0.2665	0.4189	0.6111	0.3916	0.6248	0.1831	0.419	0.2157	0.4437	0.1391	0.3785	0.2469	0.4476
11.8852	15.0	1605	10.8775	0.2088	0.4137	0.1993	0.0564	0.1788	0.3643	0.2388	0.3642	0.3879	0.2372	0.3324	0.5194	0.3625	0.5649	0.1555	0.3304	0.1954	0.4138	0.1242	0.2615	0.2062	0.3689
11.4842	16.0	1712	9.3761	0.2585	0.5013	0.2454	0.0648	0.2309	0.4104	0.2752	0.4329	0.4671	0.2841	0.4057	0.6237	0.4659	0.6676	0.2002	0.4329	0.2453	0.4701	0.1407	0.3308	0.2402	0.4342
11.1006	17.0	1819	9.2561	0.2683	0.5134	0.2582	0.0777	0.234	0.435	0.2884	0.4437	0.4887	0.2774	0.4407	0.6443	0.4587	0.6586	0.2317	0.4519	0.2363	0.4647	0.1587	0.4015	0.2559	0.4667
10.9366	18.0	1926	9.3039	0.2626	0.4996	0.251	0.0669	0.2413	0.4333	0.2889	0.4399	0.4722	0.2662	0.4241	0.6188	0.4581	0.6572	0.1891	0.4076	0.2421	0.467	0.1489	0.3692	0.2748	0.46
10.7473	19.0	2033	9.4736	0.2649	0.5138	0.2541	0.082	0.2386	0.4461	0.2883	0.4318	0.4722	0.2856	0.4252	0.6165	0.4438	0.655	0.2526	0.4519	0.222	0.4598	0.1568	0.3492	0.2494	0.4453
10.7605	20.0	2140	9.2816	0.269	0.5104	0.2442	0.0765	0.2403	0.4417	0.293	0.4501	0.4914	0.2695	0.4398	0.6531	0.4441	0.6523	0.2429	0.4582	0.2428	0.4951	0.1547	0.3862	0.2606	0.4653
10.7865	21.0	2247	9.3265	0.2757	0.5368	0.2621	0.0731	0.2484	0.4379	0.2912	0.443	0.4793	0.2896	0.441	0.6178	0.4589	0.6455	0.2726	0.481	0.2327	0.4768	0.1692	0.3354	0.245	0.4578
10.5171	22.0	2354	9.5773	0.2554	0.506	0.2458	0.0866	0.2131	0.435	0.2866	0.4385	0.4752	0.3181	0.4226	0.621	0.4187	0.6405	0.2536	0.4633	0.2051	0.4545	0.1539	0.3738	0.2457	0.444
10.7464	23.0	2461	9.4040	0.2544	0.5064	0.2414	0.0697	0.2089	0.4424	0.288	0.4357	0.4732	0.2358	0.4181	0.641	0.4118	0.6468	0.2571	0.4709	0.1966	0.4504	0.1574	0.36	0.2489	0.4378
10.5963	24.0	2568	9.1140	0.272	0.5293	0.2615	0.0838	0.2326	0.4477	0.2921	0.4473	0.4877	0.3056	0.4231	0.65	0.4342	0.6572	0.2733	0.4772	0.226	0.4808	0.1649	0.3631	0.2618	0.46
10.4877	25.0	2675	9.2811	0.2738	0.5409	0.2496	0.0811	0.2321	0.4614	0.2978	0.4528	0.4859	0.2937	0.4261	0.6431	0.4298	0.6527	0.2947	0.5025	0.2465	0.4714	0.1532	0.3662	0.2445	0.4364
10.5136	26.0	2782	9.2285	0.2809	0.5362	0.2521	0.0764	0.2341	0.4694	0.3014	0.4547	0.4955	0.2986	0.4321	0.6618	0.4342	0.6725	0.2799	0.4911	0.247	0.4701	0.1726	0.3862	0.271	0.4578
10.4462	27.0	2889	9.1017	0.2803	0.5419	0.2617	0.0914	0.2296	0.4515	0.2973	0.4663	0.5034	0.3121	0.4374	0.6604	0.4503	0.6842	0.2965	0.4899	0.2318	0.4853	0.1656	0.4077	0.2574	0.4498
10.3325	28.0	2996	9.0687	0.2849	0.5344	0.256	0.0947	0.2288	0.4799	0.3076	0.4598	0.4947	0.3034	0.4193	0.6639	0.4589	0.6752	0.2833	0.4709	0.2302	0.4754	0.1863	0.3969	0.2658	0.4551
10.3327	29.0	3103	9.1673	0.2818	0.5364	0.264	0.0932	0.2415	0.4556	0.3083	0.4606	0.4995	0.3186	0.432	0.659	0.4379	0.6784	0.2937	0.4772	0.2241	0.4647	0.18	0.4185	0.2731	0.4587
10.296	30.0	3210	9.1012	0.2813	0.5271	0.2685	0.0879	0.2399	0.4613	0.3061	0.4664	0.5014	0.2985	0.4465	0.6698	0.4438	0.6815	0.2983	0.4924	0.2305	0.4817	0.1591	0.3969	0.275	0.4547

Framework versions

Transformers 4.47.0.dev0
Pytorch 2.5.0+cu118
Datasets 2.21.0
Tokenizers 0.20.0

📄 License

This project is licensed under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご