Interview questions in Computer Vision, page 3
Translation equivariance versus invariance in CNNs
Convolution is equivariant, shifting input shifts feature maps; invariance comes only from pooling and global aggregation. Strict invariance is partial and broken by strided sampling.
Image classification versus object detection
Classification assigns one label to the whole image; detection localizes and labels multiple objects with bounding boxes and class scores.
Intersection over Union for detection
IoU is the area of overlap divided by the area of union of predicted and ground-truth boxes; a threshold decides true positives.
Mean Average Precision in object detection
AP is the area under the precision-recall curve per class; mAP averages AP over classes, and COCO also averages over IoU thresholds.
Focal Loss and class imbalance in detectors
Focal loss multiplies cross-entropy by a (1-p)^gamma factor that down-weights easy, well-classified examples so the vast easy background does not swamp the loss.
Deploying real-time detection on edge devices
Pick an efficient one-stage detector, train with augmentation, then quantize, prune, and compile to a hardware-accelerated runtime, measuring latency and accuracy tradeoffs.
Detector head losses: regression versus classification
The head splits into a classification branch using cross-entropy over classes and a regression branch using a robust Smooth L1 or IoU loss on box offsets, combined as a weighted sum.
Semantic versus instance segmentation
Semantic segmentation labels each pixel by class but merges objects of the same class; instance segmentation also separates individual objects.
U-Net architecture and its skip connections
U-Net has a contracting encoder, an expanding decoder, and skip connections that concatenate matching-resolution encoder features into the decoder to recover spatial detail lost in downsampling.
How does Mask R-CNN do instance segmentation?
Faster R-CNN backbone plus RPN, then RoIAlign and a parallel mask head predicting per-class binary masks.
How to improve coarse segmentation boundaries?
Skip connections and higher-resolution features, boundary-aware losses, and point-based or CRF refinement.
Explain panoptic segmentation and Panoptic Quality
Panoptic assigns every pixel a class and instance id over things and stuff; PQ factors into SQ, average IoU of matches, times RQ, an F1 over matched segments.
Adapting ViT for dense semantic segmentation
Reassemble patch tokens into a 2D feature map, add a decoder, and handle low resolution plus quadratic attention cost.
Sparse vs dense optical flow and Lucas-Kanade
Sparse flow tracks selected feature points, dense flow computes a vector per pixel; Lucas-Kanade solves brightness constancy in a local window assuming constant motion.
Design a tracking-by-detection tracker
Detect per frame, then associate boxes across frames by IoU or appearance using Hungarian matching, maintaining track ids.
Brightness constancy and small-motion assumptions
Brightness constancy says a point's intensity is invariant under motion; small motion lets you linearize via Taylor expansion.
Kalman filter for bounding-box tracking
State, transition, measurement models, and process plus measurement noise; predict then correct each frame. State holds box position and velocity; measurement is the detected box.
3D CNNs vs two-stream action recognition
3D CNNs learn spatiotemporal filters end to end but are heavy; two-stream splits RGB appearance and precomputed optical flow, strong but costly to compute flow.
Re-identification in multi-object tracking
Re-ID matches an object to its prior id using appearance embeddings, not just position; store track features and match re-entering detections by embedding similarity.
Self-supervised pretraining for video understanding
Define a label-free task like temporal order prediction or contrastive clip matching that forces temporal reasoning, then fine-tune on labeled action data.
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