Search Know Your ML

Search across AI research reports and briefs. Results load instantly — no JavaScript required.

Fig. 02 / Field notesFind your way to the source
A citation index with a magnifying lens locating a source.
Follow the index from a question to its original research.

22 results for “vision”

AI research reports

VGG: Very Deep Convolutional Networks

AI-generated review · 5 min read

VGG showed that stacking small 3x3 convolutions into 16-19 layer CNNs wins at ImageNet classification — the simple deep-learning blueprint behind modern vision.

vgg cnn computer-vision Read report

Mask R-CNN: Instance Segmentation

AI-generated review · 5 min read

Mask R-CNN adds a mask branch and RoIAlign to Faster R-CNN for instance segmentation — pixel-level masks, boxes and classes from one network.

instance-segmentation mask-rcnn object-detection Read report

Visualizing Convolutional Networks

AI-generated review · 5 min read

Zeiler and Fergus visualized CNN layers with deconvolution, revealing edges, textures and objects emerging layer by layer inside the network.

interpretability cnn visualization Read report

Segment Anything

AI-generated review · 6 min read

How the Segment Anything project reframed segmentation as a promptable task and released a 1-billion-mask dataset that made zero-shot segmentation practical.

segmentation computer-vision foundation-models Read report

Denoising Diffusion Probabilistic Models

AI-generated review · 8 min read

An AI-generated review of DDPM, the 2020 paper that turned diffusion models into the leading family of generative models for images.

diffusion generative-models computer-vision Read report

AI briefs

ImageNet classification with deep convolutional neural networks (AlexNet)

AI-generated brief · Cited sources below

Demonstrates that GPU-trained deep CNNs can dramatically outperform previous approaches on large-scale image classification, catalyzing the modern deep learning era in computer vision.

CLIP: learning transferable visual models from natural language supervision

AI-generated brief · Cited sources below

Trains a shared image-text embedding space with a contrastive objective on 400 million web pairs, enabling zero-shot image classification for any text-defined label set.

Sources: arXiv:2103.00020

Vision Transformer: an image is worth 16x16 words

AI-generated brief · Cited sources below

Applies a standard transformer encoder directly to image patches; with large-scale pre-training it matches or beats convolutional networks on image classification.

Sources: arXiv:2010.11929