Real Time Image based Path Tracking Algorithm.md

Real Time Image based Path Tracking Algorithm

User Interface

We have the following possible edge interface

({
    "node": "type: Node | init: node",
    "edge": "type: Edge | init: edge",
    "confidence": "type: int | init: 0",
    "last_matched_i_with_j": "type: int | init: self.i_at_matched_node - 1",
    "last_matched_j": "type: int | init: 0",
    "no_of_frames_to_match": "type: int | init: 3",
    "no_of_continuous_no_match": "type: int | init: 0",
    "edge_ended": "type: bool | init: False"
})

We achieve this interface by declaring possible edge as dictionary

Step 1

1. We open the graph_obj

graph_obj: Graph = load_graph()

2. Then we make the empty NodeEdgeRealTimeMatching object

node_and_edge_real_time_matching = NodeEdgeRealTimeMatching(graph_obj)

3. After that we run save_distinct_realtime_modified_ImgObj a. in case of live stream from a ip camera

   url = "http://192.168.43.1:8080/shot.jpg"
   folder_path_where_to_save_frames = "query_distinct_frame"
   save_distinct_realtime_modified_ImgObj(url, folder_path_where_to_save_frames, 4,check_blurry=False, ensure_min=True, livestream=True)

   b. in case you query the video from local storage

   query_video_local_url = "testData/MOV.MP4"
   folder_path_where_to_save_frames = "query_distinct_frame"
   save_distinct_realtime_modified_ImgObj(url, folder_path_where_to_save_frames, 4,
                                              check_blurry=False, ensure_min=True, livestream=True)

Goals

1. Improving and fixing bugs of chatbox
2. Adding the feature of sending multiple files and allowing receiver user to accept any number of files
3. Connecting the workflow of Anweshan and Arpit
   a. Anweshan: Before Connecting the users
   b. Arpit: After files are send by user1 and accepted by user2
   

Worflow

A. Chatbox

1. Socket.io is used in the chatbox
2. Socket Events:
   • Chat:
     • user1 & user2 connected. Message from user1 to user2
       • user1 to Server

       "message" : {
           messageValue: "<message body>"  
       }

       • Server to user1 and user2

       "message" : { 
           username: "<username1>", 
           messageValue: "<message body>",  
           timeStamp: "<timestamp>"
       }  

3. Chat messages are temporary and are not stored anywhere because it is assumed for now that connection is terminated on closing the tab

B. Accepting and Rejecting incoming FileList request

1. Suppose user1 sends the multiple files request to user2 using FileList object
2. Socket Events
   • FileList Send Request
     • user1 & user2 connected. FileList Send Request from user1 to user2
       • user1 to Server

       "fileListSendRequest" : {
           fileList: "type: FileList"
       }

       then user1 filesSendingState is set to "waiting"
       • Server to user2

       "fileListSendRequest" : {
           fileList: "type: FileList"  
       }

3. Now based on what files are accepted by user2 Socket Events:
   • File List Request Answer:
     • user2 to Server

     "fileListRequestAnswer" : {
         acceptedFilesAnswers: "type: boolean[]>"  
     }

     then user1 and user2 filesSendingState is set to "sending" and "receiving" repectively if atleast 1 file is accepted otherwise user1 and user2 filesSendingState status becomes "idle" and no file is transfered
     • Server to user1

     "fileListRequestAnswer" : {
         acceptedFilesAnswers: "type: boolean[]>"  
     }

4. acceptedFilesAnswers is an array with length(acceptedFilesAnswers) = length(fileList) that contains answer to individual files that were sent by user1. For files which has answer true, file sending process begins

Links

• https://pad.devclub.in/p/filesend