Aluode commited on
Commit
e867c14
·
verified ·
1 Parent(s): 49036e5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -70
README.md CHANGED
@@ -1,70 +1,8 @@
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.
 
1
+ language: en
2
+ license: mit
3
+ tags:pytorchvaepredictive-codingneurosciencegabor-splattingbiology
4
+ datasets:nielsr/flowers-102Neuro Splat: Artificial Visual CortexAn image is not a million pixels. It is a smooth map and a scatter of small bright wave-packets dropped where the detail lives.PerceptionLab / Antti Luode. Helsinki, June 2026.Do not hype. Do not lie. Just show.What this model isThis 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.The model is a Variational Autoencoder (VAE) trained on the Oxford Flowers-102 dataset.The Encoder acts as the top-down visual pathway, compressing an image into a latent "concept".The Decoder outputs the physical parameters (X, Y, scale, complex phase, frequency) for 512 localized Gabor atoms.The Renderer splats these wave-packets onto a canvas to create the final image via Moiré interference.The Blur is the Prior. Reality is the Sharpness.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.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).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.How to run itThis model requires the original scripts from the ArtificialCortex GitHub Repository.1. Live Cortex Perception (Webcam required)To watch the wave-packets phase-lock to your reality in real time:pip install torch torchvision numpy opencv-python
5
+ python live_cortex.py --model_path model.pt --num_packets 512
6
+ 2. Standard GenerationTo see the model hallucinate blurry priors from random latent vectors:python splat_generator.py --mode sample --resume model.pt --num_packets 512
7
+ The Honest LedgerWhat this model proves:Images can be effectively represented and learned as sparse sums of localized Gabor wave-packets.Phase-wrapping discontinuities in neural networks can be solved by outputting complex (a, b) coefficients instead of raw angles (the janus_cabbage skew medium).Predictive coding works: A top-down blurry prior can be actively sharpened by gradient descent against a live bottom-up residual (the webcam).What this model lacks: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.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.It is 2D. The current parameters map flat 2D geometries.LineageA sub-organ of the_artificial_cortex project.
8
+ 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.