yearofthewhopper

// adam ferris found it:

#import <lights>
struct PhongMaterialParameters {
vec3 emission;
vec3 ambientFactor;
vec3 diffuseFactor;
vec3 specularFactor;
float shininess;
float occlusion;

yearofthewhopper

import os
import numpy as np
import cv2

from keras.applications.imagenet_utils import preprocess_input
from keras.layers import Dense, Reshape
from keras.models import Sequential, Model, load_model
from keras.applications.resnet50 import ResNet50
from keras.optimizers import Adam

yearofthewhopper

const t = require('Time');
const D = require('Diagnostics');
const Scene = require('Scene');

const objText = Scene.root.find('2dText0');
let frame = 0;

const round = (val, precision = 1) => {
	const multiplier = Math.pow(10, precision || 0);
	return Math.round(val * multiplier) / multiplier;

yearofthewhopper

//////////////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////
// Plexus
// Collab by Ben Ursu & Dan Cronin | Afrosquared
// Spark AR Studio
// Instagram | https://www.instagram.com/a/r/?effect_id=2029175904053487

yearofthewhopper

//==============================================================================
// The following example demonstrates how to store, retrieve and remove data
// across multiple sessions.
//
// Project setup:
// - Insert two rectangles
// - Insert text
// - Position the rectangles and text so that all are visible
// - Add 'data' to 'Whitelisted keys' under the 'Persistence' Capability
// - Add the 'Tap Gesture' capability under 'Touch Gestures'

yearofthewhopper

import { Particle } from './Particle';
import { Ease } from './PFTween';
const FaceTracking = require("FaceTracking");

const HeadRotationTrigger = require("./sparkar-head-rotation-trigger.js");
const EventHelpers = require("./sparkar-event-helpers.js");

const Scene = require('Scene');
const TouchGestures = require('TouchGestures');
const Diagnostics = require('Diagnostics');

yearofthewhopper

const Scene = require('Scene')
const Animation = require('Animation')

Scene.root.findFirst('emitter0').then(obj => {
    const alphaSampler = Animation.samplers.HSVA([
        Animation.samplers.constant(1),
        Animation.samplers.constant(1),
        Animation.samplers.constant(1),
        Animation.samplers.easeInQuad(1, 0)
      ]);

yearofthewhopper

const Scene = require('Scene');
const Reactive = require('Reactive');
const TouchGestures = require('TouchGestures')


var mug_ctrl = Scene.root.find('portal')
var planeTracker = Scene.root.find('planeTracker0');


TouchGestures.onTap().subscribe(function(gesture) {

yearofthewhopper

#source https://github.com/matterport/Mask_RCNN/issues/1361

#You can 'filter out' classes you don't need after you run the detection but it doesn't prune the weights though. So, I'm not sure it would speed things up. Hope it helps.

# COCO class names
class_names = ['BG', 'person', 'bicycle', 'car', 'motorcycle', 'airplane',
               'bus', 'train', 'truck', 'boat', 'traffic light',
               'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird',
               'cat', 'dog', 'horse', 'sheep', 'cow', 'elephant', 'bear',
               'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie',

yearofthewhopper

void main() {
    vec4 textureColor = texture2D(sTexture, vTextureCoord);
    vec3 fragRGB = textureColor.rgb;
    vec3 fragHSV = rgb2hsv(fragRGB);
    float h = vHSV.x / 360.0;
    fragHSV.x *= h;
    fragHSV.yz *= vHSV.yz;
    fragHSV.x = mod(fragHSV.x, 1.0);
    fragHSV.y = mod(fragHSV.y, 1.0);
    fragHSV.z = mod(fragHSV.z, 1.0);