mhtirogla
Home
Systems
API
Pricing
About
Log in
← Back
Home
›
Systems
›
Embedded and IoT
camera-isp
camera image signal processor pipeline
smartphone ISP
System
Techniques · 4
Primitives · 24
camera-isp
Auto exposure focus white balance
media · medium: Camera 3A loops: metering-driven exposure, contrast/phase autofocus, gray-world/illuminant white balance
Image alignment registration
media · medium: Align frames via feature matching or pyramidal optical flow for burst merging and stitching
Image denoising
media · medium: Bilateral, non-local means and temporal noise reduction for sensor images
Video stabilization
media · medium: Gyro- or feature-based motion estimation, path smoothing and frame warping
Color space conversion
graphics · low: Convert pixels between transfer curves (sRGB, linear light), primaries (sRGB, Display P3, XYZ) and Y'CbCr, using tables for the curve and 3x3 matrices for the rest.
›
Convolution image filters
graphics · low: Separable 2D convolution kernels for blur, sharpen (unsharp mask) and edges with border handling
Image pyramids
graphics · low: Gaussian/Laplacian pyramids and tiled multi-resolution images for zoom and blending
Image warping remap
graphics · low: Resampling with lookup maps: resize, lens undistortion, homographies, bilinear/bicubic
Dma transfers
embedded · low: DMA controllers: descriptors, circular buffers, cache coherence for zero-CPU data movement
Deterministic fixed-point math
game-networking · low: Bit-identical simulation math across CPUs, compilers and builds: integer fixed-point formats, the float traps (FMA, x87, fast-math, libm), deterministic trig, sqrt and RNG.
›
Tone mapping (HDR to display range)
graphics · low: Compress scene-referred linear light to a display's range in three steps: expose, apply a curve to luminance, then encode (sRGB, PQ or HLG) and dither.
›
SIMD programming
hardware-os · low: Process 4-16 lanes per instruction with AVX2, AVX-512 or NEON via Rust std::arch, with runtime dispatch and a portable fallback.
›
Demosaicing
media · low: Reconstruct full RGB from Bayer sensor data (bilinear, edge-aware, gradient-corrected)
Line buffer streaming
media · low: Stream image processing with a few line buffers instead of full frames (hardware ISPs)
Convolutional networks
ml-training · low: Layers that slide a small shared kernel over a grid, giving translation equivariance and few parameters; still the most compute-efficient choice for many vision tasks.
›
Weight quantization
ml-training · low: Store weights as 8/4-bit integers with per-group scales; decode is memory-bound, so 4-bit weights give about 3-4x tokens/s and fit 4x larger models.
›
PID controller
robotics · low: Drive a measured value to a setpoint with three terms on the error: proportional, integral, derivative. The default feedback loop for motors, heaters, drones and gimbals.
›
Peak detection
signal-processing · low: Robust peak and event detection in noisy signals with adaptive thresholds and refractory periods
Homogeneous transforms
geometry · low: 4x4 affine and projective matrices, homogeneous coordinates, view and projection, normal matrices
Feature detection orb
robotics · low: FAST corners with oriented BRIEF descriptors and Hamming matching for tracking and relocalization
+ 8 more primitives
System
Techniques · 4
Primitives · 24
camera-isp
Auto exposure focus white balance
media · medium: Camera 3A loops: metering-driven exposure, contrast/phase autofocus, gray-world/illuminant white balance
Image alignment registration
media · medium: Align frames via feature matching or pyramidal optical flow for burst merging and stitching
Image denoising
media · medium: Bilateral, non-local means and temporal noise reduction for sensor images
Video stabilization
media · medium: Gyro- or feature-based motion estimation, path smoothing and frame warping
Color space conversion
graphics · low: Convert pixels between transfer curves (sRGB, linear light), primaries (sRGB, Display P3, XYZ) and Y'CbCr, using tables for the curve and 3x3 matrices for the rest.
›
Convolution image filters
graphics · low: Separable 2D convolution kernels for blur, sharpen (unsharp mask) and edges with border handling
Image pyramids
graphics · low: Gaussian/Laplacian pyramids and tiled multi-resolution images for zoom and blending
Image warping remap
graphics · low: Resampling with lookup maps: resize, lens undistortion, homographies, bilinear/bicubic
Dma transfers
embedded · low: DMA controllers: descriptors, circular buffers, cache coherence for zero-CPU data movement
Deterministic fixed-point math
game-networking · low: Bit-identical simulation math across CPUs, compilers and builds: integer fixed-point formats, the float traps (FMA, x87, fast-math, libm), deterministic trig, sqrt and RNG.
›
Tone mapping (HDR to display range)
graphics · low: Compress scene-referred linear light to a display's range in three steps: expose, apply a curve to luminance, then encode (sRGB, PQ or HLG) and dither.
›
SIMD programming
hardware-os · low: Process 4-16 lanes per instruction with AVX2, AVX-512 or NEON via Rust std::arch, with runtime dispatch and a portable fallback.
›
Demosaicing
media · low: Reconstruct full RGB from Bayer sensor data (bilinear, edge-aware, gradient-corrected)
Line buffer streaming
media · low: Stream image processing with a few line buffers instead of full frames (hardware ISPs)
Convolutional networks
ml-training · low: Layers that slide a small shared kernel over a grid, giving translation equivariance and few parameters; still the most compute-efficient choice for many vision tasks.
›
Weight quantization
ml-training · low: Store weights as 8/4-bit integers with per-group scales; decode is memory-bound, so 4-bit weights give about 3-4x tokens/s and fit 4x larger models.
›
PID controller
robotics · low: Drive a measured value to a setpoint with three terms on the error: proportional, integral, derivative. The default feedback loop for motors, heaters, drones and gimbals.
›
Peak detection
signal-processing · low: Robust peak and event detection in noisy signals with adaptive thresholds and refractory periods
Homogeneous transforms
geometry · low: 4x4 affine and projective matrices, homogeneous coordinates, view and projection, normal matrices
Feature detection orb
robotics · low: FAST corners with oriented BRIEF descriptors and Hamming matching for tracking and relocalization
Ransac
robotics · low: Random sample consensus for robust model fitting with outliers (homographies, planes, poses)
Quaternion rotations
geometry · low: Unit quaternions as 3D rotations: 4 floats, cheap composition, no gimbal lock, plus the conversions, interpolation choices and numerical care needed in practice.
›
IIR and FIR filters
signal-processing · low: Two ways to shape a sampled signal's spectrum: FIR (finite taps, always stable, exact linear phase) or IIR (feedback, a few coefficients, cheap but phase-distorting).
›
Chroma subsampling
media · low: 4:2:0/4:2:2 colour plane downsampling and siting
Low precision number formats
numerics · low: Bf16, fp16, fp8 (E4M3, E5M2), MX block formats and int4: ranges, rounding, scaling
LU and Cholesky factorization
numerics · low: Factor A = PLU (any square matrix, partial pivoting) or A = LLᵀ (SPD) once in O(n³), then solve each right-hand side in O(n²).
›
PRNGs (PCG, xoshiro)
numerics · low: Small, fast, statistically strong non-cryptographic generators: PCG32 and xoshiro256**, with seeding, parallel streams and unbiased ranges.
›
Biquad filters
signal-processing · low: Second-order IIR sections (RBJ cookbook): low/high pass, peaking EQ; stable cascades
Copy for AI
Similar systems
Share the most building blocks with this one.
raster-image-editor
19 %
machine-vision-inspection
19 %
sequencer-instrument
18 %