Aluode commited on
Commit
49036e5
·
verified ·
1 Parent(s): b016709

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +36 -54
README.md CHANGED
@@ -1,88 +1,70 @@
1
- language: en
2
- license: mit
3
- tags:
4
 
5
- pytorch
 
 
 
 
 
 
6
 
7
- vae
8
 
9
- predictive-coding
10
 
11
- neuroscience
12
-
13
- gabor-splatting
14
-
15
- biology
16
- datasets:
17
-
18
- nielsr/flowers-102
19
-
20
- Neuro Splat: Artificial Visual Cortex
21
-
22
- An image is not a million pixels. It is a smooth map and a scatter of small bright wave-packets dropped where the detail lives.
23
-
24
- PerceptionLab / Antti Luode. Helsinki, June 2026.
25
 
26
  Do not hype. Do not lie. Just show.
27
 
28
- What this model is
29
 
30
- This repository contains the weights (model.pt) and the architecture for a differentiable Gabor wave-packet splatter. It is not a standard pixel-generating AI. It is an experimental representation of the mammalian primary visual cortex (V1), built to test predictive coding and phase-amplitude coupling.
31
 
32
  The model is a Variational Autoencoder (VAE) trained on the Oxford Flowers-102 dataset.
33
 
34
- The Encoder acts as the top-down visual pathway, compressing an image into a latent "concept".
35
-
36
- The Decoder outputs the physical parameters (X, Y, scale, complex phase, frequency) for 512 localized Gabor atoms.
37
 
38
- The Renderer splats these wave-packets onto a canvas to create the final image via Moiré interference.
39
 
40
- The Blur is the Prior. Reality is the Sharpness.
41
-
42
- If you ask this model to generate a flower from scratch, it will produce a soft, blurry, watercolor-like blob. This is the expected, biologically accurate behavior.
43
 
44
  When an AI (or a brain) lacks raw input, the safest mathematical guess is to average all ambiguous possibilities. It produces the low-frequency "gist" (the Theta carrier).
45
 
46
  To see the model actually work, you must give it eyes. By hooking this model up to a live webcam, the blurry top-down prior collides with raw, bottom-up reality. The live video supplies the exact phase, the wave-packets lock into alignment, and the blurry gist snaps into razor-sharp perception.
47
 
48
- How to run it
49
 
50
- This model requires the original scripts from the ArtificialCortex GitHub Repository.
51
 
52
- 1. Live Cortex Perception (Webcam required)
53
 
54
  To watch the wave-packets phase-lock to your reality in real time:
55
 
56
- pip install torch torchvision numpy opencv-python
57
- python live_cortex.py --model_path model.pt --num_packets 512
58
-
59
 
60
- 2. Standard Generation
61
 
62
  To see the model hallucinate blurry priors from random latent vectors:
63
 
64
- python splat_generator.py --mode sample --resume model.pt --num_packets 512
65
-
66
-
67
- The Honest Ledger
68
-
69
- What this model proves:
70
-
71
- Images can be effectively represented and learned as sparse sums of localized Gabor wave-packets.
72
-
73
- Phase-wrapping discontinuities in neural networks can be solved by outputting complex (a, b) coefficients instead of raw angles (the janus_cabbage skew medium).
74
 
75
- Predictive coding works: A top-down blurry prior can be actively sharpened by gradient descent against a live bottom-up residual (the webcam).
76
 
77
- What this model lacks:
78
 
79
- It is not a photorealistic generator. It is a VAE with an amortized latent space. It competes on biological representation, not on Generative AI benchmarks against Diffusion models.
 
 
80
 
81
- The "Floaters": When subjected to high-stress visual input (like a webcam feed of a room instead of a flower), the optimizer often orphans wave-packets, resulting in bright, isolated dots ("floaters" or "phosphenes"). This occurs because the model currently lacks lateral inhibitory coordination between packets.
82
 
83
- It is 2D. The current parameters map flat 2D geometries.
 
 
84
 
85
- Lineage
86
 
87
- A sub-organ of the_artificial_cortex project.
88
- The generator dreamed a blurry flower because it had nothing to look at. Open its eyes and reality supplies the phase the dream could not; the packets lock, the gist sharpens, and what they cannot yet coordinate, they see as stars.
 
1
+ language: en license: mit tags:
 
 
2
 
3
+ * pytorch
4
+ * vae
5
+ * predictive-coding
6
+ * neuroscience
7
+ * gabor-splatting
8
+ * biology datasets:
9
+ * nielsr/flowers-102
10
 
11
+ # **Neuro Splat: Artificial Visual Cortex**
12
 
13
+ ### **An image is not a million pixels. It is a smooth map and a scatter of small bright wave-packets dropped where the detail lives.**
14
 
15
+ **PerceptionLab / Antti Luode. Helsinki, June 2026\.**
 
 
 
 
 
 
 
 
 
 
 
 
 
16
 
17
  Do not hype. Do not lie. Just show.
18
 
19
+ ## **What this model is**
20
 
21
+ This repository contains the weights (model.pt) and the architecture for a **differentiable Gabor wave-packet splatter**. It is not a standard pixel-generating AI. It is an experimental representation of the mammalian primary visual cortex (V1), built to test predictive coding and phase-amplitude coupling.
22
 
23
  The model is a Variational Autoencoder (VAE) trained on the Oxford Flowers-102 dataset.
24
 
25
+ * **The Encoder** acts as the top-down visual pathway, compressing an image into a latent "concept".
26
+ * **The Decoder** outputs the physical parameters (X, Y, scale, complex phase, frequency) for 512 localized Gabor atoms.
27
+ * **The Renderer** splats these wave-packets onto a canvas to create the final image via Moiré interference.
28
 
29
+ ## **The Blur is the Prior. Reality is the Sharpness.**
30
 
31
+ If you ask this model to generate a flower from scratch, it will produce a soft, blurry, watercolor-like blob. **This is the expected, biologically accurate behavior.**
 
 
32
 
33
  When an AI (or a brain) lacks raw input, the safest mathematical guess is to average all ambiguous possibilities. It produces the low-frequency "gist" (the Theta carrier).
34
 
35
  To see the model actually work, you must give it eyes. By hooking this model up to a live webcam, the blurry top-down prior collides with raw, bottom-up reality. The live video supplies the exact phase, the wave-packets lock into alignment, and the blurry gist snaps into razor-sharp perception.
36
 
37
+ ## **How to run it**
38
 
39
+ This model requires the original scripts from the [ArtificialCortex GitHub Repository](https://github.com/anttiluode/ArtificialCortex/tree/main/the_splat).
40
 
41
+ ### **1\. Live Cortex Perception (Webcam required)**
42
 
43
  To watch the wave-packets phase-lock to your reality in real time:
44
 
45
+ pip install torch torchvision numpy opencv-python
46
+ python live\_cortex.py \--model\_path model.pt \--num\_packets 512
 
47
 
48
+ ### **2\. Standard Generation**
49
 
50
  To see the model hallucinate blurry priors from random latent vectors:
51
 
52
+ python splat\_generator.py \--mode sample \--resume model.pt \--num\_packets 512
 
 
 
 
 
 
 
 
 
53
 
54
+ ## **The Honest Ledger**
55
 
56
+ **What this model proves:**
57
 
58
+ * Images can be effectively represented and learned as sparse sums of localized Gabor wave-packets.
59
+ * Phase-wrapping discontinuities in neural networks can be solved by outputting complex (a, b) coefficients instead of raw angles (the janus\_cabbage skew medium).
60
+ * Predictive coding works: A top-down blurry prior can be actively sharpened by gradient descent against a live bottom-up residual (the webcam).
61
 
62
+ **What this model lacks:**
63
 
64
+ * **It is not a photorealistic generator.** It is a VAE with an amortized latent space. It competes on biological representation, not on Generative AI benchmarks against Diffusion models.
65
+ * **The "Floaters":** When subjected to high-stress visual input (like a webcam feed of a room instead of a flower), the optimizer often orphans wave-packets, resulting in bright, isolated dots ("floaters" or "phosphenes"). This occurs because the model currently lacks lateral inhibitory coordination between packets.
66
+ * **It is 2D.** The current parameters map flat 2D geometries.
67
 
68
+ ## **Lineage**
69
 
70
+ A sub-organ of the\_artificial\_cortex project. The generator dreamed a blurry flower because it had nothing to look at. Open its eyes and reality supplies the phase the dream could not; the packets lock, the gist sharpens, and what they cannot yet coordinate, they see as stars.