@srsergio/taptapp-ar 1.0.88 → 1.0.89

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +13 -8
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -29,10 +29,12 @@
29
29
 
30
30
  ## 🌟 Key Features
31
31
 
32
- - 🖼️ **Hyper-Fast Compiler**: Pure JavaScript compiler that generates `.taar` files in **< 3s**.
32
+ - 🎭 **Non-Rigid Surface Tracking**: Supports curved and deformable surfaces using **Delaunay Meshes** and **Mass-Spring Relaxation**.
33
+ - 🚀 **Hyper-Fast Compiler**: Pure JavaScript compiler that generates `.taar` files in **< 3s**.
33
34
  - ⚡ **No TensorFlow Dependency**: No TFJS at all. Works natively in any JS environment (Node, Browser, Workers).
34
35
  - 🧬 **Fourier Positional Encoding**: Uses high-frequency sine/cosine mappings (GPT-style) for neural-like spatial consistency.
35
36
  - 🚀 **Protocol V7 (Moonshot)**:
37
+ - **Delaunay Triangular Grid**: Adaptive mesh that tracks surface deformations.
36
38
  - **16-bit Fourier Signatures**: Spatial ADN embedded in every feature for harmonic matching.
37
39
  - **4-bit Packed Tracking Data**: Grayscale images are compressed to 4-bit depth, slashing file size.
38
40
  - **64-bit LSH Descriptors**: Optimized Locality Sensitive Hashing for descriptors.
@@ -54,9 +56,9 @@ npm install @srsergio/taptapp-ar
54
56
 
55
57
  | Metric | Official MindAR | TapTapp AR V7 | Improvement |
56
58
  | :--- | :--- | :--- | :--- |
57
- | **Compilation Time** | ~23.50s | **~2.61s** | 🚀 **~9x Faster** |
58
- | **Output Size (.taar)** | ~770 KB | **~50 KB** | 📉 **93% Smaller** |
59
- | **Descriptor Format** | 84-byte Float | **64-bit LSH** | 🧠 **Massive Data Saving** |
59
+ | **Compilation Time** | ~23.50s | **~0.93s** | 🚀 **~25x Faster** |
60
+ | **Output Size (.taar)** | ~770 KB | **~338 KB** | 📉 **56% Smaller** |
61
+ | **Descriptor Format** | 84-byte Float | **128-bit LSH** | 🧠 **Massive Data Saving** |
60
62
  | **Tracking Data** | 8-bit Gray | **4-bit Packed** | 📦 **50% Data Saving** |
61
63
  | **Dependency Size** | ~20MB (TFJS) | **< 100KB** | 📦 **99% Smaller Bundle** |
62
64
 
@@ -68,11 +70,11 @@ The latest version has been rigorously tested with an adaptive stress test (`rob
68
70
 
69
71
  | Metric | Result | Description |
70
72
  | :--- | :--- | :--- |
71
- | **Pass Rate** | **96.3%** | High success rate across resolutions. |
72
- | **Drift Tolerance** | **< 15%** | Validated geometrically against ground truth metadata. |
73
+ | **Pass Rate** | **93.5%** | High success rate across resolutions (202/216). |
74
+ | **Drift Tolerance** | **< 10%** | Validated geometrically against ground truth metadata. |
73
75
  | **Tracking Precision** | **Float32** | Full 32-bit precision for optical flow tracking. |
74
- | **Detection Time** | **~21ms** | Ultra-fast initial detection on standard CPU. |
75
- | **Total Pipeline** | **~64ms** | Complete loop (Detect + Match + Track + Validate). |
76
+ | **Detection Time** | **< 20ms** | Ultra-fast initial detection on standard CPU. |
77
+ | **Total Pipeline** | **~45ms** | Complete loop (Detect + Match + Track + Validate). |
76
78
 
77
79
  ---
78
80
 
@@ -258,6 +260,8 @@ ar.stop();
258
260
  ## 🏗️ Protocol V7 (Moonshot Packed Format)
259
261
  TapTapp AR uses a proprietary **Moonshot Vision Codec** that is significantly more efficient than standard AR formats.
260
262
 
263
+ - **Non-Rigid Surface Tracking**: Replaces the standard rigid homography with a dynamic **Delaunay Mesh**. This allows the tracker to follow the curvature of posters on cylinders, t-shirts, or slightly bent magazines.
264
+ - **Mass-Spring Relaxation**: The tracking mesh is optimized using physical relaxation, minimizing L2 distance between predicted and tracked points while maintaining topological rigidity.
261
265
  - **Fourier Positional Encoding**: Maps 2D coordinates into a 16-dimensional frequency space. This creates a "Neural Consistency Check" that filters out noise and motion blur by checking for harmonic spatial agreement.
262
266
  - **4-bit Packed Tracking Data**: Image data used for optical flow is compressed to 4-bit depth.
263
267
  - **64-bit LSH Fingerprinting**: Feature descriptors are compressed to just 8 bytes using LSH.
@@ -271,3 +275,4 @@ TapTapp AR uses a proprietary **Moonshot Vision Codec** that is significantly mo
271
275
  MIT © [srsergiolazaro](https://github.com/srsergiolazaro)
272
276
 
273
277
  Based on the core research of MindAR, but completely re-written for high-performance binary processing and JS-only execution.
278
+
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@srsergio/taptapp-ar",
3
- "version": "1.0.88",
3
+ "version": "1.0.89",
4
4
  "description": "Ultra-fast Augmented Reality (AR) SDK for Node.js and Browser. Image tracking with 100% pure JavaScript, zero-dependencies, and high-performance compilation.",
5
5
  "keywords": [
6
6
  "augmented reality",