毕业设计&课设-ECO跟踪器的Matlab实现。.zip

### News: Check our [pytracking](https://github.com/visionml/pytracking) repository, which includes a PyTorch-based implementation of ECO and our most recent trackers [ATOM (CVPR 2019)](https://arxiv.org/pdf/1811.07628.pdf) and [DiMP (ICCV 2019)](https://arxiv.org/pdf/1904.07220.pdf). # ECO Matlab implementation of the Efficient Convolution Operator (ECO) tracker. ## Publication Details about the tracker can be found in the CVPR 2017 paper: [Martin Danelljan](https://martin-danelljan.github.io/), Goutam Bhat, Fahad Khan, Michael Felsberg. <a href="https://arxiv.org/abs/1611.09224">ECO: Efficient Convolution Operators for Tracking</a>. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. Please cite the above publication if you use the code or compare with the ECO tracker in your work. Bibtex entry: @InProceedings{DanelljanCVPR2017, Title = {ECO: Efficient Convolution Operators for Tracking}, Author = {Danelljan, Martin and Bhat, Goutam and Shahbaz Khan, Fahad and Felsberg, Michael}, Booktitle = {CVPR}, Year = {2017} } ## Project Webpage http://www.cvl.isy.liu.se/research/objrec/visualtracking/ecotrack/index.html ## Contact [Martin Danelljan](https://martin-danelljan.github.io/) ## Installation ### Using git clone 1. Clone the GIT repository: $ git clone https://github.com/martin-danelljan/ECO.git 2. Clone the submodules. In the repository directory, run the commands: $ git submodule init $ git submodule update 3. Start Matlab and navigate to the repository. Run the install script: |>> install 4. Run the demo script to test the tracker: |>> demo_ECO Note: This package requires matconvnet [1], if you want to use deep CNN features, and PDollar Toolbox [2], if you want to use HOG features. Both these externals are included as git submodules and should be installed by following step 2. above. ### Without using git You could also downlad and install without using git. This is however not recommented since it will be harder to incorporate updates and you will not get the correct versions of matconvnet and PDollar Toolbox. 1. Download ZIP file from https://github.com/martin-danelljan/ECO and unpack it somewhere. 2. Download matconvnet ZIP file from https://github.com/vlfeat/matconvnet and unpack it in the external_libs/matconvnet/ folder of the repository. Download PDollar Toolbox ZIP file from https://github.com/pdollar/toolbox and unpack it in the external_libs/pdollar_toolbox/ folder of the repository. Lastly, perform steps 3. and 4. above. ## Description and Instructions ### Runfiles The files in the runfiles/ directory are uset to set parameters and run the tracker. You can create your own runfiles by copying an existing one and then play around with different parameters and feature combinations. These runfiles are included: * OTB_DEEP_settings.m - Settings in the paper used for the deep feature version of ECO on the OTB, UAV123 and TempleColor datasets. * OTB_HC_settings.m - Settings in the paper used for the hand-crafted feature version of ECO (ECO-HC) on the OTB, UAV123 and TempleColor datasets. * VOT2016_DEEP_settings.m - Settings in the paper used for the deep feature version of ECO on the VOT2016 dataset. * VOT2016_HC_settings.m - Settings in the paper used for the hand-crafted feature version of ECO (ECO-HC) on the VOT2016 dataset. * testing_ECO.m - Demo file which contains the same settings as OTB_DEEP_settings.m by default. * testing_ECO_gpu.m - Same as testing_ECO, but uses GPU. * testing_ECO_HC.m - Demo file which contains the same settings as OTB_HC_settings.m by default. * SRDCF_settings.m - Settings that roughly correspond to our previous SRDCF tracker [4] from 2015. The main difference is the optimization method. Tracking performance on the OTB-2015 dataset is shown bellow for different settings. For comparison, results of our previous trackers C-COT [3], SRDCF [4], DeepSRDCF [5] and DSST [6] are included. <img src="https://github.com/martin-danelljan/ECO/blob/master/result_plots/OTB-2015_succsess_plot.png" alt="Could not display image" height=400 width=500> ### Features This package includes a quite general framework for feature extraction. You can easily incorporate your own features in the same manner by adding a corresponding "get_featureX.m" function. Currently, four types of features are included: 1. Deep CNN features. It uses matconvnet [1], which is included as a git submodule in external_libs/matconvnet/. The imagenet-vgg-m-2048 network available at http://www.vlfeat.org/matconvnet/pretrained/ was used in the ECCV paper. You can use other networks, by placing them in the feature_extraction/networks/ folder. 2. HOG features. It uses the PDollar Toolbox [2], which is included as a git submodule in external_libs/pdollar_toolbox/. 3. Lookup table features. These are implemented as a lookup table that directly maps an RGB or grayscale value to a feature vector. Currently, Color Names [8] and Intensity Channels [9] are included. 4. Colorspace features. Currently grayscale and RGB are implemented. The tracker supports almost any combination of features. Currently, the only limitation is that you can only use deep features from a single network (but you can use several different layers from the same network). Each feature has its own parameter settings. You can set the cell size for each non-CNN feature independently. ECO does not assume the same cell size for all feature channels. For the CNN features, you can control the cell size by setting an additional down-sampling factor for each layer. See the runfile testing.m for examples of how to integrate different features. You can uncomment several features at once in the params.t_features cell array. ### fDSST Scale Filter This reposetery also includes an implementation of the optimized scale filter, which was originally proposed in the fDSST [7]. It is here used in the ECO-HC version of the tracker for speeding-up the scale estimation. ### GPU Support GPU support is activated by setting the parameter "params.use_gpu = true" in the runfile. This requires MatConvNet to be compiled with GPU support. If the install script fails, please visit http://www.vlfeat.org/matconvnet/install/ for instructions. ### Integration Into OTB It should be easy to integrate the tracker into the Online Tracking Benchmark [10]. The runfiles supports the OTB interface, so you just have to copy and rename the runfile you want to use and then add the necessary paths (see setup_paths.m). ### Integration Into VOT To integrate the tracker into the Visual Object Tracking (VOT) challenge toolkit [11], check the VOT_integration folder. Copy the configuration file to your VOT workspace and set the path to the ECO reposetory inside it. The tracker now supports the trax protocol, which is necessary for VOT2017 version of the toolkit. ### Raw Results All raw result files used in our CVPR 2017 paper can be found at the project webpage: http://www.cvl.isy.liu.se/research/objrec/visualtracking/ecotrack/index.html ### Why Does the Result Change? Tracking performance may vary slightly on different machines and whether GPU support is activated. This is due to small numerical effects which can accumulate over time (all trackers are affected by this). Generally, this only affects the performance marginally. More significant changes are rare, but can occur in some videos. ## Code References *[visionml/pytracking](https://github.com/visionml/pytracking)*: Python (PyTorch) implementation of ECO and general tracking library containing official implementation of our latest trackers [ATOM](https://arxiv.org/pdf/1811.07628.pdf) and [DiMP](https://arxiv.org/pdf/1904.07220.pdf) *[rockkingjy/OpenTracker](https://github.com/rockkingjy/OpenTracker)*: C++ Implementation of ECO and o