MCPcopy Create free account

hub / github.com/JeffWang987/EgoVid / functions

Functions54 in github.com/JeffWang987/EgoVid

⨍ Functions 54◇ Types & classes 1
↓ 5 callersMethodextrinsic2pyramid
(self, extrinsic, color_map='red', hw_ratio=9/16, base_xval=1, zval=3)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:25
↓ 4 callersMethodsavefig
(self, save_path)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:60
↓ 2 callersFunctionadd_legend
(img, text='hello', position=(0, 0), colour=[255, 255, 255], size=14)
processor/egovid-low-level-action(cam_control)/imu_vis.py:48
↓ 2 callersFunctionfilter_noise
(acce, fps, noise_cutoff_freq=1.0)
processor/egovid-low-level-action(cam_control)/imu_vis.py:72
↓ 2 callersFunctionintegrate_imu_data
(imu_data, initial_velocity, fps)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:75
↓ 2 callersFunctionremove_gravity
Parameters: acce: N x 3 numpy array of acceleration values (xyz) in m/s^2 fps: The sampling rate of the IMU signals in Hz
processor/egovid-low-level-action(cam_control)/imu_vis.py:58
↓ 2 callersFunctiontransform_extrinsics
(colmap_extrinsics)
processor/egovid-low-level-action(cam_control)/colmap_pose_vis.py:74
↓ 1 callersFunctionOptions
()
processor/egovid-metadata/CLIP/clipsim.py:70
↓ 1 callersFunctionOptions
()
processor/egovid-metadata/EgoVideo/backbone/egovideo_clean.py:12
↓ 1 callersFunctionOptions
()
processor/egovid-high-level-action(caption)/qwen_response.py:13
↓ 1 callersFunctionOptions
()
processor/egovid-high-level-action(caption)/llava_response.py:22
↓ 1 callersFunctionOptions
()
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:214
↓ 1 callersFunctioncart_to_polar_manual
(x, y)
processor/egovid-metadata/Optical-flow/optical-flow-score.py:39
↓ 1 callersMethodcolorbar
(self, max_frame_length)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:51
↓ 1 callersFunctionconnect_point_trajectory
(args, image_dir, output_dir, skip_exists=False, keep_intermediate=False, raft1=None, raft2=None)
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:37
↓ 1 callersFunctionext_interpolation
(paths, ratio=5, wirte_txt=False, write_path=None)
processor/egovid-low-level-action(cam_control)/colmap_pose_vis.py:23
↓ 1 callersFunctionext_interpolation
(paths, ratio=5, wirte_txt=False, write_path=None)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:25
↓ 1 callersFunctionfuse_results
(results: list)
processor/egovid-metadata/DOVER/egovid_dover.py:23
↓ 1 callersFunctionget_args
()
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:63
↓ 1 callersFunctionget_c2w
(w2cs, transform_matrix, relative_c2w)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:83
↓ 1 callersFunctionget_clip_model
(clip_model_name="ViT-L-14", pretrained='openai', device='cpu')
processor/egovid-metadata/CLIP/clipsim.py:62
↓ 1 callersFunctionget_img_img_similarity
(img1, img2, clip_model, img_preprocess, device='cpu')
processor/egovid-metadata/CLIP/clipsim.py:35
↓ 1 callersFunctionget_optical_flow
(model, image1, image2, device='cpu')
processor/egovid-metadata/Optical-flow/optical-flow-score.py:38
↓ 1 callersFunctionget_response
(video_path, tokenizer, model, image_processor, input_ids, attention_masks, stopping_criteria, stop_str)
processor/egovid-high-level-action(caption)/llava_response.py:108
↓ 1 callersFunctionget_text_img_similarity
(text, img, clip_model, img_preprocess, text_tokenizer, device='cpu')
processor/egovid-metadata/CLIP/clipsim.py:17
↓ 1 callersFunctionintegrate_imu_rot
(init_rot_matrix, angular_velocities, fps)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:194
↓ 1 callersFunctionjacobian_of_f
状态转移函数的雅可比矩阵
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:174
↓ 1 callersFunctionjacobian_of_h
观测模型的雅可比矩阵, 因为是直接从colmap中读取的, 认为可以直接观测到
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:183
↓ 1 callersFunctionload_model
Run inference on ActivityNet QA DataSet using the Video-ChatGPT model. Args: args: Command-line arguments.
processor/egovid-high-level-action(caption)/llava_response.py:62
↓ 1 callersFunctionload_optical_flow_model
(model_path, device='cpu')
processor/egovid-metadata/Optical-flow/optical-flow-score.py:16
↓ 1 callersFunctionload_video
(video_path, args)
processor/egovid-high-level-action(caption)/llava_response.py:48
↓ 1 callersFunctionmake_contour
(img, colour=[0, 0, 0], double_line=False)
processor/egovid-low-level-action(cam_control)/imu_vis.py:27
↓ 1 callersFunctionmotion_segmentation
(args, image_dir, output_dir, traj_dir, skip_exists=False, keep_intermediate=False, motion_model=None)
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:60
↓ 1 callersFunctionobservation_model
(x)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:124
↓ 1 callersFunctionoptimize_params
(colmap_pose, imu_acc, fps)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:95
↓ 1 callersFunctionparse_args
()
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:131
↓ 1 callersFunctionparticlesfm
Inputs: - img_dir: str - The folder containing input images - output_dir: str - The workspace directory
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:107
↓ 1 callersFunctionquat_derivative
Compute quaternion derivative given the current quaternion (q) and 3D angular rate vector (omega).
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:139
↓ 1 callersFunctionquat_multiply
Multiply two quaternions.
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:127
↓ 1 callersFunctionread_points3d_bin
(file_path)
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:30
↓ 1 callersFunctionread_video
(path, eval_frames=None)
processor/egovid-metadata/CLIP/clipsim.py:53
↓ 1 callersFunctionread_video
(path, eval_frames=None)
processor/egovid-metadata/EgoVideo/backbone/egovideo_clean.py:31
↓ 1 callersFunctionread_video
(path, eval_frames=None)
processor/egovid-low-level-action(cam_control)/imu_vis.py:16
↓ 1 callersFunctionread_video
(path)
processor/egovid-low-level-action(cam_control)/colmap_pose_vis.py:16
↓ 1 callersFunctionread_video
(path)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:19
↓ 1 callersFunctionread_video
(path)
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:176
↓ 1 callersFunctionsfm_reconstruction
(args, image_dir, output_dir, traj_dir, skip_exists=False, keep_intermediate=False)
processor/egovid-low-level-action(cam_control)/particle-sfm/run_particlesfm_videoin_ego4djson.py:86
↓ 1 callersMethodshow
(self)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:56
↓ 1 callersFunctionstate_transition
(x, u, dt)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:126
Method__init__
(self, xlim, ylim, zlim)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:12
Methodcustomize_legend
(self, list_label)
processor/egovid-low-level-action(cam_control)/colmap_traj_vis.py:43
Functiongaussian_rescale
(pr)
processor/egovid-metadata/DOVER/egovid_dover.py:30
Functionresiduals
(params, colmap_pose, imu_acc, fps)
processor/egovid-low-level-action(cam_control)/kalman_imu_colmap.py:97
Functionuniform_rescale
(pr)
processor/egovid-metadata/DOVER/egovid_dover.py:36