Video yükleniyor...
Video Yüklenemedi
It's always been this easy to edit the normals into a more 2D style and i've only just figured it out 🥲 no more concave eye shadows 🙏 Alt+N > Copy Vectors > Paste Vectors #blender
39,161 görüntüleme • 2 yıl önce •via X (Twitter)
8 Yorum
How does this look for the end product? It looked like it made the brows more floaty
The eyes will look flatter instead of caverns. But only when shader is set up for it
I've been constantly learning photoshop for over 10 years and I still find new things, I expect to be learning blender for my whole life 😅could you explain a bit more the principle behind the vectors and the normals and how that relates to shading for us rookies in the back pls?
It allows you to force Impossible shading so you can create more 2d looking lighting instead of being forced by the shape of your model :) for example the eyes will look flat instead of concave
There something similar that I do with fur, where I use a Data Transfer modifier and use a vertex group gradient to blend in the fur into the main mesh.
If you want an addon for helping to take even more full control of the normals, the addon on GitHub called "Abnormal" recently just got updated for the newest version of blender. It adds whole new layout just entirely for editing the vertex and face vectors on your mesh!
This will be very useful for the fucked up shading on my lashes, thanks!
There’s also a free add on (which I can’t remember the name rn, I’ll add it when I figure it out) that can help with normals for a npr look where you select your mesh > select the normals style > press enter and done lol
![[CLIP] by Hand ✍️ The CLIP (Contrastive Language–Image Pre-training) model, a groundbreaking work by OpenAI, redefines the intersection of computer vision and natural language processing. It is the basis of all the multi-modal foundation models we see today. How does CLIP work? Goal: 🟨 Learn a shared embedding space for text and image [1] Given ↳ A mini batch of 3 text-image pairs ↳ OpenAI used 400 million text-image pairs to train its original CLIP model. Process 1st pair: "big table" [2] 🟪 Text → 2 Vectors (3D) ↳ Look up word embedding vectors using word2vec. [3] 🟩 Image → 2 Vectors (4D) ↳ Divide the image into two patches. ↳ Flatten each patch [4] Process other pairs ↳ Repeat [2]-[3] [5] 🟪 Text Encoder & 🟩 Image Encoder ↳ Encode input vectors into feature vectors ↳ Here, both encoders are simple one layer perceptron (linear + ReLU) ↳ In practice, the encoders are usually transformer models. [6] 🟪 🟩 Mean Pooling: 2 → 1 vector ↳ Average 2 feature vectors into a single vector by averaging across the columns ↳ The goal is to have one vector to represent each image or text [7] 🟪 🟩 -> 🟨 Projection ↳ Note that the text and image feature vectors from the encoders have different dimensions (3D vs. 4D). ↳ Use a linear layer to project image and text vectors to a 2D shared embedding space. 🏋️ Contrastive Pre-training 🏋️ [8] Prepare for MatMul ↳ Copy text vectors (T1,T2,T3) ↳ Copy the transpose of image vectors (I1,I2,I3) ↳ They are all in the 2D shared embedding space. [9] 🟦 MatMul ↳ Multiply T and I matrices. ↳ This is equivalent to taking dot product between every pair of image and text vectors. ↳ The purpose is to use dot product to estimate the similarity between a pair of image-text. [10] 🟦 Softmax: e^x ↳ Raise e to the power of the number in each cell ↳ To simplify hand calculation, we approximate e^□ with 3^□. [11] 🟦 Softmax: ∑ ↳ Sum each row for 🟩 image→🟪 text ↳ Sum each column for 🟪 text→ 🟩 image [12] 🟦 Softmax: 1 / sum ↳ Divide each element by the column sum to obtain a similarity matrix for 🟪 text→🟩 image ↳ Divide each element by the row sum to obtain a similarity matrix for 🟩 image→🟪 text [13] 🟥 Loss Gradients ↳ The "Targets" for the similarity matrices are Identity Matrices. ↳ Why? If I and T come from the same pair (i=j), we want the highest value, which is 1, and 0 otherwise. ↳ Apply the simple equation of [Similarity - Target] to compute gradients of for both directions. ↳ Why so simple? Because when Softmax and Cross-Entropy Loss are used together, the math magically works out that way. ↳ These gradients kick off the backpropagation process to update weights and biases of the encoders and projection layers (red borders).](https://image.24vids.com/tw-1801370086152016034/ext_tw_video_thumb/1801368762937131008/pu/img/GuRh5lcJ6M7HsrkY.jpg)
