光学学报, 2018, 38 (10): 1015002, 网络出版: 2019-05-09
稳健的双模型自适应切换实时跟踪算法 下载: 797次
Robust Real-Time Visual Tracking via Dual Model Adaptive Switching
图 & 表
图 2. 单层自适应特征与模型切换跟踪部分结果。(a) Skiing; (b) Walking2; (c) Freeman1; (d) Tiger1
Fig. 2. Partial tracking results of single-layer adaptive features and model switching. (a) Skiing; (b) Walking2; (c) Freeman1; (d) Tiger1
图 3. 目标在不同卷积层的特征信息表达。(a)目标图像;VGG-19的(b) pool4和(c)conv5-3输出卷积特征
Fig. 3. Expression of object characteristics in different convolution layer. (a) Target image; output convolution feature of (b) pool4 and (c) conv5-3 with VGG-19
图 4. 自适应卷积特征选择(a)前、(b)后对比图
Fig. 4. Comparison of images (a) before and (b) after adaptive convolution features selection
图 5. Freeman1视频跟踪过程的PSR与最大响应值变化曲线。(a) PSR变化曲线;(b)最大响应值变化曲线
Fig. 5. PSR and maximum response curves of Freeman1 video tracking process. (a) PSR curves; (b) maximum response curves
图 6. 10种算法的(a)精度和(b)成功率曲线
Fig. 6. (a) Precision and (b) success rate plots of ten algorithms
图 7. 面对不同挑战时各算法精确度曲线。(a) IV; (b) OPR; (c) SV; (d) OCC; (e) DEF; (f) MB; (g) FM; (h) IPR; (i) OV; (j) BC; (k) LR
Fig. 7. Precision plots of ten algorithms on different challenges. (a) IV; (b) OPR; (c) SV; (d) OCC; (e) DEF; (f) MB; (g) FM; (h) IPR; (i) OV; (j) BC; (k) LR
图 8. 面对不同挑战时各算法成功率值。(a) IV; (b) OPR; (c) SV; (d) OCC; (e) DEF; (f) MB; (g) FM; (h) IPR; (i) OV; (j) BC; (k) LR
Fig. 8. Success rate plots of ten algorithms on different challenges. (a) IV; (b) OPR; (c) SV; (d) OCC; (e) DEF; (f) MB; (g) FM; (h) IPR; (i) OV; (j) BC; (k) LR
表 1特征不同应用方式的跟踪结果
Table1. Tracking results of different features applications
|
表 2测试视频的属性
Table2. Properties of test videos
|
熊昌镇, 车满强, 王润玲, 卢颜. 稳健的双模型自适应切换实时跟踪算法[J]. 光学学报, 2018, 38(10): 1015002. Changzhen Xiong, Manqiang Che, Runling Wang, Yan Lu. Robust Real-Time Visual Tracking via Dual Model Adaptive Switching[J]. Acta Optica Sinica, 2018, 38(10): 1015002.