More in Computer Vision — page 3
Compare Gray World and White Patch white balance.
WHAT IT TESTS: classic color constancy methods. OUTLINE: Gray World assumes average scene color is gray, White Patch assumes the brightest pixel is white, both fail on dominant colors or clipping; learning predicts illuminant from data.
How does smartphone Portrait Mode produce bokeh?
WHAT IT TESTS: depth estimation plus synthetic rendering. OUTLINE: estimate per-pixel depth via dual-pixel or stereo or learning, segment the subject, then apply depth-dependent blur.
Outline the classic image stitching pipeline.
WHAT IT TESTS: feature-based image stitching. OUTLINE: detect and match features like SIFT, estimate a homography with RANSAC, warp and blend with multiband or feathering.
How do you build an HDR image from bracketed exposures?
WHAT IT TESTS: HDR imaging pipeline basics. OUTLINE: align frames, recover the camera response function, merge to a linear radiance map weighted by exposure, then tone map for display.
How do BYOL and Barlow Twins avoid representation collapse?
WHAT IT TESTS: self-supervised learning and collapse avoidance. OUTLINE: collapse is embeddings shrinking to a constant or low-rank subspace; BYOL uses predictor plus momentum target plus stop-gradient, Barlow Twins decorrelates feature dimensions.
How does MAML's inner and outer loop work?
WHAT IT TESTS: meta-learning and bi-level optimization. OUTLINE: inner loop does task-specific gradient steps from shared init, outer loop updates the init for fast adaptability via second-order gradients.
Prototypical Networks for few-shot classification
WHAT IT TESTS: metric-based few-shot learning. OUTLINE: an encoder embeds support examples, each class prototype is the mean embedding of its support examples, and a query is classified by nearest prototype using a distance like Euclidean via softmax.
Contrastive self-supervised learning with SimCLR
WHAT IT TESTS: contrastive learning mechanics. OUTLINE: two augmentations of one image form a positive pair, other images in the batch are negatives; an encoder plus projection head and the NT-Xent loss pull positives together and push negatives apart.
Leveraging unlabeled data with limited labels
WHAT IT TESTS: semi-supervised and self-supervised strategy. OUTLINE: pretrain a representation on the million unlabeled images via self-supervision, then fine-tune on the 1,000 labels; or use pseudo-labeling and consistency regularization.
Transfer learning from ResNet50 on small data
WHAT IT TESTS: applying transfer learning. OUTLINE: replace the final classification head with one sized to your classes, freeze the pretrained convolutional backbone as a feature extractor, train the new head, then optionally fine-tune top blocks at a low…
Formulating a multi-step robot manipulation task
WHAT IT TESTS: end-to-end robot RL formulation. OUTLINE: perception detects and localizes the mug, action space spans navigation and manipulation, and a reward shaped over subgoals (reach, grasp, transport, place) with sparse final success guides learning.
NeRF limitations and advances for robotics
WHAT IT TESTS: practical limits of NeRF. OUTLINE: original NeRF is slow to train and render, per-scene, static, and needs many calibrated views; address speed with explicit grids or Gaussian splatting, dynamics with time-conditioned fields, and scale with…
Camera intrinsics, extrinsics, and the essential matrix
WHAT IT TESTS: epipolar geometry fundamentals. OUTLINE: intrinsics map camera coords to pixels, extrinsics are camera pose in the world; the essential matrix relates normalized points across two views, encoding relative rotation and translation up to scale…
Designing a baseline Visual Question Answering model
WHAT IT TESTS: multimodal baseline design. OUTLINE: encode the image with a CNN, encode the question with an RNN or embedding, fuse the two vectors, and classify over a fixed answer vocabulary.
Semantic, instance, and panoptic segmentation
WHAT IT TESTS: distinguishing segmentation paradigms. OUTLINE: semantic labels every pixel by class without separating objects; instance separates individual objects but may skip background; panoptic unifies both, labeling stuff and distinct thing instances.
Designing a high-resolution photorealistic face generator
WHAT IT TESTS: system design for high-res faces. OUTLINE: weigh StyleGAN's fast, controllable style-based synthesis against diffusion's diversity and stable training; handle scale via progressive or multi-resolution synthesis; protect diversity to avoid mode…
DDPM versus DDIM sampling trade-offs
WHAT IT TESTS: stochastic versus deterministic sampling. OUTLINE: DDPM is a stochastic Markov chain needing many steps; DDIM is a non-Markovian, deterministic sampler that skips steps for far faster inference and reproducible, invertible latents, trading a…
Classifier-free guidance in diffusion models
WHAT IT TESTS: how guidance improves conditioning. OUTLINE: train one model jointly on conditional and dropped-condition inputs; at inference extrapolate from unconditional toward conditional prediction via a guidance scale, sharpening prompt adherence…
Unpaired image translation with CycleGAN
WHAT IT TESTS: unpaired translation design. OUTLINE: CycleGAN uses two generators and two discriminators with a cycle-consistency loss that forces translating to the other domain and back to reconstruct the input, removing the need for paired data.
How text prompts guide Stable Diffusion
WHAT IT TESTS: the text-conditioning pipeline. OUTLINE: a frozen text encoder turns the prompt into token embeddings, which feed the U-Net via cross-attention at each denoising step so the prompt steers generation; classifier-free guidance amplifies the…