Image Pixels, Channels, and Color Spaces
Read an image as an array: inspect RGB channels, grayscale and HSV, and understand coordinates, data types and tensor layouts before preprocessing.
Related
4 min read · beginner
Images, recognition, detection, segmentation, visual representations, camera geometry, and motion.
Start with image arrays, then understand how sampling and local filters change the information they contain.
Starting knowledge: Basic arithmetic; no neural-network background required
Select a concept. Explore its connections, or open the full explanation.
Previewing Pixels and color spaces.
Inspect the image representation, change its sampling grid, then use neighborhoods to smooth an image or detect changes.
From pixels to local structure
Read RGB channels, coordinates and tensor layouts, and compare grayscale with HSV.
Read conceptUsed in: connections 1 to 3 of 5.
Return here after learning boxes or masks to see how a transform must preserve the relationship between an image and its labels.
Lines show suggested reading within a branch. Connections offer optional ways to explore; they are not a checklist of prerequisites.
Read an image as an array: inspect RGB channels, grayscale and HSV, and understand coordinates, data types and tensor layouts before preprocessing.
4 min read · beginner
Compare nearest-neighbor and bilinear image resizing, see aliasing when downsampling, and understand aspect ratio, letterboxing and mask interpolation.
4 min read · beginner
Inspect Gaussian smoothing and Sobel filters as local weighted sums, understand signed gradients and Canny edges, and connect fixed filters to CNNs.
4 min read · beginner
Transform images and annotations together, inspect crop and flip geometry, and choose augmentation assumptions that preserve the meaning of the task.
4 min read · beginner
Follow pixels into image features, compare spatial maps with pooled vectors, and see why image encoders expose different outputs for recognition and localization.
6 min read · beginner
Turn an image into a sequence: count patches, flatten pixels, apply a shared learned projection, and separate token count from embedding width.
6 min read · beginner
Interactive guide to convolution in CNNs: visualize sliding windows, kernels, stride, padding, and feature detection with step-by-step demos.
10 min read
Understand receptive fields in CNNs: how convolutional layers expand their field of view and the gap between theoretical and effective receptive fields.
7 min read
Understand dilated (atrous) convolutions: how dilation rates expand receptive fields without extra parameters and how to avoid gridding artifacts.
12 min read
Learn how skip connections and residual learning enable training of very deep neural networks. Understand the ResNet revolution with interactive visualizations.
9 min read
Follow an image through convolution, ReLU, pooling and class scores. Inspect a transparent toy CNN and learn what classification can and cannot predict.
4 min read · beginner
Understand where self-supervised targets come from, how masked prediction creates a learning task, and why predicting pixels differs from predicting learned features.
7 min read · beginner
Follow how a learned target encoder supplies feature targets, how stop-gradient separates backward paths, and how exponential moving averages update its parameters.
6 min read · beginner
Move a pretrained encoder to a downstream task: trace frozen features, head training, partial and full fine-tuning, learning rates, and BatchNorm state.
9 min read · intermediate
Evaluate frozen image features with nearest neighbors and linear probes, separate training from validation and test data, and understand what each protocol can establish.
9 min read · beginner
Convert between xyxy, xywh and center-size boxes, normalize coordinates correctly, and avoid mistakes when resizing, clipping or flipping detection labels.
4 min read · beginner
Understand object detection from image features to boxes, classes and scores. Compare one-stage and two-stage pipelines, training and duplicate handling.
3 min read · beginner
Compare anchor-based vs anchor-free object detection: Faster R-CNN and RetinaNet anchors vs FCOS and CenterNet point-based methods.
12 min read · intermediate
Understanding Non-Maximum Suppression algorithms for object detection post-processing, from greedy NMS to soft variants
12 min read · intermediate
Follow spatial image features into DETR decoder slots, separating learned query embeddings, attention queries, changing content and object predictions.
6 min read · beginner
Learn how DETR assigns unordered object targets to prediction slots, why greedy choices can fail, and how matching differs from training loss and evaluation.
7 min read · beginner
Understanding end-to-end object detection with transformers, from DETR's object queries to bipartite matching and attention-based localization
15 min read · advanced
Follow a query from a spatial reference through learned offsets, bilinear feature sampling and weighted aggregation, with exact non-square coordinate arithmetic.
8 min read · intermediate
Understand cross-entropy loss for classification: interactive demos of binary and multi-class CE, the -log(p) curve, softmax gradients, and focal loss.
15 min read
Learn focal loss for deep learning: down-weight easy examples, focus on hard ones. Interactive demos of gamma, alpha balancing, and RetinaNet.
12 min read
Understanding region-based feature extraction for object detection, from quantized pooling to sub-pixel alignment and adaptive sampling
12 min read · intermediate
Calculate box intersection and union, explore how position and size change IoU, and distinguish overlap used for evaluation, NMS and localization losses.
4 min read · beginner
Match detections to objects, inspect precision–recall curves, calculate average precision, and distinguish confidence thresholds from IoU thresholds.
4 min read · beginner
Learn how normalized box coordinates, L1 loss, IoU, and Generalized IoU guide object localization, including disjoint boxes and containment limits.
5 min read · beginner
Compare semantic, instance and panoptic label maps, inspect individual pixels, and understand dense predictions, mask IoU, Dice and boundary errors.
4 min read · beginner
Follow coarse feature maps into aligned pixel predictions: nearest and bilinear interpolation, transposed convolution, output shapes, skip fusion, and resizing logits before selecting labels.
7 min read · intermediate
Learn how Feature Pyramid Networks build multi-scale feature representations through top-down pathways and lateral connections for robust object detection.
10 min read · intermediate
Learning where to fuse multi-scale features with per-pixel, per-level fusion weights. ASFF challenges FPN's uniform fusion assumption.
15 min read · intermediate
Understanding how neural architecture search discovers optimal feature pyramid architectures that outperform hand-designed alternatives
15 min read · advanced
Follow a 3D world point through camera extrinsics, depth division and pixel intrinsics, then see why one image point cannot determine its depth.
7 min read · beginner
Follow image points between frames, distinguish forward flow from backward sampling, and understand brightness constancy, occlusion and the aperture problem.
7 min read · beginner
Compare edge, frequency and entropy heuristics for selecting a fixed budget of image patches, and distinguish token counts from accuracy and measured speed.
8 min read · advanced