(191212) Data Visualization: Color changing using rendered material

<Result> <Code> import rhinoscriptsyntax as rs import Rhino import scriptcontext import System.Drawing def exportData(): # Generating sensor points pts = setSensorLocation() # Make csv file file_object = open("points1.csv", "w") # import sensor value file_data = open("values.csv", "r") data = file_data.readlines() valList = [] for index in range(0, len(data)): valNum = float(data[index]) valList.append(valNum) maxNum = max(valList) … Continue reading (191212) Data Visualization: Color changing using rendered material

(191210) Data visualization with point: How do we visualize data using points?

<Result> <Code> import rhinoscriptsyntax as rs import random as rd def exportData(): pts = rs.GetObjects("Select") file_object = open("points1.csv", "w") #valList = [] i = 0 for pt in pts: coord = rs.PointCoordinates(pt) tempVal = genVal() #dataLine = str(coord) + " ,\n" dataLine = str(coord) dataLine = dataLine + " ," + str(tempVal) + "\n" #valList.append(tempVal) … Continue reading (191210) Data visualization with point: How do we visualize data using points?

(191025) ACADIA – Day 2

25, Oct 2019 (UTC-06:00 Central time) <Session 5 : Material research> All of material researches focused on novel module to construct building elements. The novel designed modules mostly created by using 3D printing system. (1) Polybrick 2 뼈의 구조 중 canacellous bone(뼈의 구조 중 구멍이 송송 뚫려있는 부분)을 모사하여 조적조 구조를 만들 수 있는 재료를 … Continue reading (191025) ACADIA – Day 2

(191024) ACADIA – Day 1

24, Oct 2019 (UTC-06:00 Central time) The place where the presentations are mainly proceeded. 1. Session 1 : AI / Deep learning / Data (1) Towards an autonomous architecture They tried to construct the concept of non-static and intelligent architecture. They generate tensegrity-based building forms by using reinforcement learning. (2) A plane of thrones They … Continue reading (191024) ACADIA – Day 1

(191007) Extracting XYZ coordinates + normalizing XYZ data

<Code> 1 Extracting XYZ def extractXYZ(): file_input = open("label1.pts", "r") # Read existing data file_output = open("label1_xyz.pts", "w") # Write and save only coorninate data dataline = file_input.readlines() #print(len(dataline)) # number of line #print(dataline[0]) # values in each line listAll = [] for i in range(0, len(dataline)): list_str = dataline[i].split() # string to list # … Continue reading (191007) Extracting XYZ coordinates + normalizing XYZ data

(191004) Training – inference : How to train by using different data with same labels

(-ing) 0) 우선적으로 inference 할 때 testing dataset이 정확히 어떻게 설정되는지를 알아야 할 것 같다. 그 뒤에 1) 앞에 세 개만 남기고 라이노로 -> threecor.py 2) reduce density -> reduce.py 3) divide point cloud by building module -> manually 아마 메쉬랩으로 먼저 importing그 뒤에 rhino -> editting 4) 각각의 빌딩 모듈과 그 주위의 것들 -> … Continue reading (191004) Training – inference : How to train by using different data with same labels

(190930) 3D Semantic segmentation: How do we perform training with point cloud and their corresponding data?

<Objective> Training a model for point cloud semantic segmentation with our preprocessed dataset files: link (Preprocessd data is too big so just pretrained model files are included) <Result> Fig 1.1. inference result. Fig 1.2. Segmented point cloud file by exporting to .ply extension & importing to Rhino.(*We cannot directly open ply file in Rhino. I … Continue reading (190930) 3D Semantic segmentation: How do we perform training with point cloud and their corresponding data?

(190927) Normalizing point data

<Objective> Mapping point data to (-1 ,1)^3 files: link <Before> ... 6.5406213630381 55.110140522295 5.14619020078717 6.38238963824233 54.6751469833632 5.09043611785057 9.66267167996784 53.8237395988369 5.25916710277026 9.73812172794146 53.2868717371414 5.25022971671459 7.1054251905191 38.2644542956424 5.31910125886565 7.0135011626237 37.1817786076376 5.3314482132195 7.43287343589282 36.6527543534698 5.37239936769243 6.93071468788321 35.9802793359437 5.3365470066215 7.17623156836794 34.1945842388266 5.33038598350655 7.20191223986512 31.6342752297926 5.39023567865158 6.76679551260713 31.1099425380205 5.3536527427704 ... <After> ... -0.21434 -0.03315 -0.32328 -0.21711 -0.04078 -0.32425 -0.15956 -0.05572 … Continue reading (190927) Normalizing point data