------------------------------------- Finding Action Tubes, CVPR 2015 Source code by Georgia Gkioxari (contact: gkioxari@eecs.berkeley.edu) ------------------------------------- The pipeline described in Finding Action Tubes (CVPR, 2015) consists of multiple steps. For simplicity, we break them down to independent procedures. 0a. In startup.m, add the paths and resolve the dependencies 0b. Selective search boxes need to be stored in the following format /ss_dir/motion/action/video/0000f.mat Video frames need to be stored in the following format /img_dir/action/video/0000f.png A. Optical flow computation compute_OF/compute_flow.m for a pair of images computes flow as described in the paper. im1, im2: input images optical flow images need to be stored in the format /flow_dir/action/video/0000f.png B. Motion saliency motion_saliency/get_motion_salient_boxes.m for each frame, prunes boxes (e.g. from Selective Search) based the optical flow signal within each box. annot: set of videos and actions, jhmdb_annot.mat ss_dir: directory containing the boxes flow_dir: directory containing the optical flow images (as computed by A.) C. Extract fc7 features extract_features/rcnn_cache_fc7_features_jhmdb.m extracts fc7 features. type: 'spatial' or 'motion' net_def_file: prototxts, models/jhmdb/extract_fc7.prototxt. Same for any type net_file: models, use pretrained models for JHMDB as provided in the project page. output_dir: cache directory, the features are cached in output_dir/type/action/video/frame.mat. D. Train SVM models train_svm/train_jhmdb.m trains SVM models, one for each action annot: ground truth information and boxes (after pruning), jhmdb_motion_sal_annot.mat feat_dir: directory with cached features save_dir: cache directory E. Action Tubes train_svm/compute_tubes.m scores and links detections to create the final action tubes annot: source of boxes, jhmdb_motion_sal_annot.mat rcnn_model: the models as computed by train_jhmdb.m F. Precomputed tubes test_tubes/ tubes for all three splits of J-HMDB and UCF sports test_tubes/UCFsports_benchmark/ AUC and ROC numbers for UCFSports and plots (see ipython notebook) G. Evaluate/ROC curves evaluate/get_ROC_curve_JHMDB.m computes ROC and AUC for JHMDB annot: ground truth annotation (annot_jhdmb.mat) tubes: tubes on the test set actions: list of actions iou_thresh: threshold for intersection over union draw: true to draw the curves (For UCF sports the same function was used with some small adjustments regarding the format of the data) --------------------------------------------------------------------------------------------------------------- G. Training spatial-CNN and motion-CNN To train the networks you need to do the following: 1. Compute the optical flow as in A. 2. Create window_train(val).txt with the window data (similar to R-CNN detection) 3. Use Caffe to train (train prototxt is given in models/jhmdb/train.prototxt), and initialize with the proper model 4. In the case of motion-CNN, you need to make two changes in window_data_layer.cpp a. The image mean is for all channels 128 (instead of the image mean provided) b. During training, when flipping of the input image occurs the flow in x needs to also change sign