Everything in Computer Vision, page 3
Compare Gray World and White Patch white balance.
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?
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.
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?
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?
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?
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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…
We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.
See open roles