[WIP] first version of model cards for bodyPose, bodySegmentation, handMesh…

content/blog/cosa-nyu-ml-tools/index.md

@@ -1,5 +1,5 @@
 ---
-templateKey: #blog-post
+templateKey: blog-post
 title: COSA x NYU Machine Learning Tools for Creative Coding
 author: ml5.js
 description: Join us at ITP for an informal series of talks and workshops exploring open-source machine learning tools for creative coding, presented in partnership with the Clinic for Open Source Arts (COSA)!

content/model-card/bodypose/index.md

@@ -0,0 +1,69 @@
+---
+templateKey: blog-post # <-- Uncomment this so that this post can be included in the blog list
+title: BodyPose Model Card
+author: ml5.js
+description: Friendly Machine Learning for the Web
+keywords: bias, model card, BodyPose
+image: "./images/_thumb.jpg"
+externalLink: (link)
+date: "2025-03-14" # YYYY-MM-DD or YYYY-MM-DD to YYYY-MM-DD or YYYY-MM-DD, YYYY-MM-DD, YYYY-MM-DD
+tags:
+  - BodyPose
+featured: true
+---
+BodyPose is developed leveraging TensorFlow's [MoveNet](https://www.tensorflow.org/hub/tutorials/movenet#:~:text=MoveNet%20is%20an%20ultra%20fast,known%20as%20Lightning%20and%20Thunder) and [BlazePose](https://ai.google.dev/edge/mediapipe/solutions/vision/pose_landmarker) models.
+
+______
+## MoveNet
+MoveNet was trained on [two datasets](https://storage.googleapis.com/movenet/MoveNet.SinglePose%20Model%20Card.pdf): 
+
+**COCO Keypoint Dataset Training Set 2017**
+- Date created: **2017**
+- Size: **28K images** 
+- How the data was collected: “In-the-wild images with diverse scenes, instance sizes, and occlusions.” The original dataset of 64K images was distilled to the final 28K to only include three or less people per image.
+- Bias:
+    * According to the public [model card](https://storage.googleapis.com/movenet/MoveNet.SinglePose%20Model%20Card.pdf), the qualitative analysis shows that although the dataset has a 3:1 male to female ratio, favors young and light skinned individuals, the models is stated to perform “fairly” (< 5% performance differences between most categories).
+        * Categories of evaluation: 
+            * Male / Female (gender)
+            * Young / Middle-age / Old (age)
+            * Darker / Medium/ Lighter (skin tone)
+    * There has been a fair amount of [research](https://medium.com/@rxtang/mitigating-gender-bias-in-captioning-systems-5a956e1e0d6d#:~:text=COCO%20dataset%20has%20an%20imbalanced,the%20bias%20learned%20by%20models) about the COCO Dataset. Most show that the dataset has numerous biases occurring due to underrepresentation of certain demographics. 
+
+**Active Dataset Training Set**
+- Date created: **2017-2021** ([assuming](https://blog.tensorflow.org/2021/05/next-generation-pose-detection-with-movenet-and-tensorflowjs.html))
+- Size: **23.5k images**
+- How the data was collected: “Images sampled from **YouTube fitness videos** which capture people exercising (e.g. HIIT, weight-lifting, etc.), stretching, or dancing. It contains diverse poses and motion with more motion blur and self-occlusions.”
+- Bias:
+    * According to the model card, the models are stated to perform “fairly” (< 5% performance differences between all categories).
+        * Categories of evaluation: 
+            * Male / Female (gender)
+            * Young / Middle-age / Old (age) 
+            * Darker / Medium/ Lighter (skin tone)
+    * The Active Single Person Image set, unlike COCO dataset, is not public, hence there is no additional research conducted to evaluate the fairness. 
+
+As stated, fitness videos uploaded to YouTube were used to assemble this internal Google dataset. Only in [2024](https://support.google.com/youtube/thread/313644973/third-party-ai-trainability-on-youtube?hl=en), Google [has provided](https://support.google.com/youtube/answer/15509945?hl=en) creators the opportunity to opt-out from Google using their videos for their AI/ML research.
+
+___
+## BlazePose
+BlazePose’s [research paper](https://arxiv.org/pdf/2006.10204) and [model card](https://drive.google.com/file/d/10WlcTvrQnR_R2TdTmKw0nkyRLqrwNkWU/preview)
+- Date created: **2020-2021 (assuming)**
+- Size: **80K**
+- How the data was collected: Not stated in the original research paper. The model card asserts: “This model was trained and evaluated on images, including consented images (30K), of people using a mobile AR application captured with smartphone cameras in various “in-the-wild” conditions. The majority of training images (85K) capture a wide range of fitness poses.” 
+- Bias:
+    * According to the model card, the models are stated to perform “fairly”.
+        * Categories of evaluation: 
+            * 14 subregions 
+            * Male / Female (gender)
+            * 6 skin tones
+        * Evaluation results:
+            * Subregions (14): difference in confidence between average and worst performing regions of 4.8% for the heavy, 4.8% for the full and 6.5% for the light model.
+            * Gender: difference in confidence is 1.1% for the heavy model, 2.2% for the full model and 3.1% for the lite model.
+            * Skin tones: difference in confidence between worst and best performing categories is 5.7% for the heavy model, 7.0% for the full model and 7.3% for the lite model. 
+
+There is no additional research conducted to evaluate the fairness.
+There is no specific information on how the **consent** was obtained to get the images. 
+
+
+____
+
+#### Please submit any feedback/information you belive would be useful regarding this model [here](https://forms.gle/BPG44g3cJywSKjde6).

content/model-card/bodysegmentation/index.md

@@ -0,0 +1,39 @@
+---
+templateKey: blog-post # <-- Uncomment this so that this post can be included in the blog list
+title: BodySegmentation Model Card
+author: ml5.js
+description: Friendly Machine Learning for the Web
+keywords: bias, model card, BodySegmentation
+image: "./images/_thumb.jpg"
+externalLink: (link)
+date: "2025-03-14" # YYYY-MM-DD or YYYY-MM-DD to YYYY-MM-DD or YYYY-MM-DD, YYYY-MM-DD, YYYY-MM-DD
+tags:
+  - BodySegmentation
+featured: true
+---
+The ml5.js BodySegmentation provides two models, **SelfieSegmentation** and **BodyPix**: 
+
+______
+## SelfieSegmentation
+
+**MediaPipe Selfie Segmentation [Model Card](https://storage.googleapis.com/mediapipe-assets/Model%20Card%20MediaPipe%20Selfie%20Segmentation.pdf)**
+- Date created: **2021**
+- Size: **Not stated** 
+- How the data was collected: “This model was trained and evaluated on images, including consented images of people using a mobile AR application captured with smartphone cameras in various “in-the-wild” conditions.” 
+- Bias:
+    * Categories of evaluation:
+        * 17 demographical subregions
+        * 6 skin tones
+        * Male / Female (gender)
+    * Evalutation results:
+        * Subregions: Difference in confidence between average and worst performing regions of 1.11% for the general model, and 1.28% for the landscape model, lower than the criteria.
+        * Gender: Differences in confidence are 1.6% for the general model and 0.6% for the landscape model.
+        * Skin tone: Difference in confidence between worst and best performing
+
+____
+## BodyPix
+This [short article](https://medium.com/tensorflow/introducing-bodypix-real-time-person-segmentation-in-the-browser-with-tensorflow-js-f1948126c2a0) is the only information on BodyPix that we have found. 
+
+____
+
+#### Please submit any feedback/information you belive would be useful regarding this model [here](https://forms.gle/BPG44g3cJywSKjde6).

content/model-card/facemesh/index.md

@@ -0,0 +1,63 @@
+---
+templateKey: blog-post # <-- Uncomment this so that this post can be included in the blog list
+title: Facemesh Model Card
+author: ml5.js
+description: Friendly Machine Learning for the Web
+keywords: bias, model card, HandPose
+image: "./images/_thumb.jpg"
+externalLink: (link)
+date: "2025-03-14" # YYYY-MM-DD or YYYY-MM-DD to YYYY-MM-DD or YYYY-MM-DD, YYYY-MM-DD, YYYY-MM-DD
+tags:
+  - HandPose
+featured: true
+---
+
+## MediaPipe Face Mesh 
+
+**These are the [Model Card](https://drive.google.com/file/d/1sv4sSb9BSNVZhLzxXJ0jBv9DqD-4jnAz/view?pli=1), the [Research](https://arxiv.org/pdf/1907.06724), and the [Research Blog](https://sites.google.com/view/perception-cv4arvr/facemesh)**
+- Date created: **2018**
+- Size: **Not stated** 
+- How the data was collected: “All dataset images were captured on a diverse set of smartphone cameras, both front- and back-facing. All images were captured in a real-world environment with different light, noise and motion conditions via an AR (Augmented Reality) application.”
+
+- Bias:
+  * According to the model card, the models are stated to perform “well” across most groups.  
+    *  Categories of evaluation: 
+        * 17 geographic subregions
+        * 6 skin tones
+        * Male / Female (gender)
+    * Evaluation results:
+        * Subregions: Difference in confidence between best and worst performing regions of 0.9% for the tracking mode and 1.56% for the reacquisition mode. 
+        * Genders: Difference in confidence is 0.02% for the tracking mode and 0.1% for the reacquisition mode. 
+        * Skin tones: Difference in confidence is 0.24% for tracking mode and 1.12% for the reacquisition mode. 
+  * There is no additional research conducted to evaluate the fairness.
+
+
+
+  * There is no additional research conducted to evaluate the fairness.
+
+_____
+## MediaPipe Attention Mesh
+**This is the [Model Card](https://drive.google.com/file/d/1tV7EJb3XgMS7FwOErTgLU1ZocYyNmwlf/preview)**
+- Date created: **2020**
+- Size: **30K (assuming)** 
+- How the data was collected: “All dataset images were captured on a diverse set of smartphone cameras, both front- and back-facing. All images were captured in a real-world environment with different light, noise and motion conditions via an AR (Augmented Reality) application.”
+
+- Bias:
+  * According to the model card, the models are stated to perform “well” across most groups.
+
+    *  Categories of evaluation: 
+        * 17 geographic subregions
+        * 6 skin tones
+        * Male / Female (gender)
+    * Evaluation results:
+        * Subregions: Difference in confidence between best and worst performing regions of 1.22% for the tracking mode and 1.27% for the reacquisition mode. 
+        * Gender: Difference in confidence is 0.01% for the tracking mode and 0.03% for the reacquisition mode. 
+        * Skin tones: Difference in confidence is 0.54% for tracking mode and 0.88% for the reacquisition mode. 
+
+- Potential bias:
+  * Potential biases in the model which may result in drastic confidence differences if evaluated: May have difficulty with facial accessories like glasses or cultural headwear.
+
+____
+
+#### Please submit any feedback/information you belive would be useful regarding this model [here](https://forms.gle/BPG44g3cJywSKjde6).
+

content/model-card/handpose/index.md

@@ -0,0 +1,29 @@
+---
+templateKey: blog-post # <-- Uncomment this so that this post can be included in the blog list
+title: HandPose Model Card
+author: ml5.js
+description: Friendly Machine Learning for the Web
+keywords: bias, model card, HandPose
+image: "./images/_thumb.jpg"
+externalLink: (link)
+date: "2025-03-14" # YYYY-MM-DD or YYYY-MM-DD to YYYY-MM-DD or YYYY-MM-DD, YYYY-MM-DD, YYYY-MM-DD
+tags:
+  - HandPose
+featured: true
+---
+## Hand Detection / Tracking 
+
+**Hand Detection/Tracking [Model Card](https://drive.google.com/file/d/1sv4sSb9BSNVZhLzxXJ0jBv9DqD-4jnAz/view?pli=1)**
+- Date created: **2021** (assuming)
+- Size: **Not stated** 
+- How the data was collected: “This model was trained and evaluated on images of people using a mobile AR application captured with smartphone cameras in various “in-the-wild” conditions.”
+
+- Bias:
+  * No evaluation has been conducted on this model. No access to the dataset was provided.  
+  * As stated on the model card: “as with many human sensing tools, performance may vary across skin tones, gender, age, and potentially other sensitive demographic characteristics.”
+  * There is no additional research conducted to evaluate the fairness.
+
+____
+
+#### Please submit any feedback/information you belive would be useful regarding this model [here](https://forms.gle/BPG44g3cJywSKjde6).
+