powerboxesrs 0.3.0

Utility functions to manipulate and compute metrics on boxes
Documentation

CI Coverage Crates.io

PowerBoxesrs

Powerboxes is a package containing utility functions for transforming bounding boxes and computing metrics.

Installation

cargo add powerboxesrs

Functions available

Note: all functions expect boxes in xyxy format (top left and bottom right corners), except box conversion and rotated box functions.

All core functions have a _slice variant operating on flat &[N] slices. ndarray wrappers are available behind the ndarray feature (enabled by default).

Box Transformations and utilities

  • box_areas: Compute the area of list of boxes
  • box_convert: Convert a box from one format to another. Supported formats are xyxy, xywh, cxcywh
  • remove_small_boxes: Remove boxes with area smaller than a threshold
  • masks_to_boxes: Convert a mask to a list of boxes

Box Metrics

  • iou_distance: Compute the intersection over union matrix of two sets of boxes
  • parallel_iou_distance: Compute the intersection over union matrix of two sets of boxes in parallel
  • giou_distance: Compute the generalized intersection over union matrix of two sets of boxes
  • parallel_giou_distance: Compute the generalized intersection over union matrix of two sets of boxes in parallel
  • diou_distance: Compute the distance intersection over union matrix of two sets of boxes
  • tiou_distance: Compute the tracking intersection over union matrix of two sets of boxes

Rotated Box Metrics

Rotated boxes use (cx, cy, w, h, angle) format where angle is in degrees.

  • rotated_iou_distance: Compute IoU distance for rotated boxes
  • rotated_giou_distance: Compute GIoU distance for rotated boxes
  • rotated_tiou_distance: Compute tracking IoU distance for rotated boxes

Box NMS

  • nms: Non-maximum suppression, returns the indices of the boxes to keep
  • rtree_nms: Non-maximum suppression using an R-tree for sub-quadratic complexity

Drawing

  • draw_boxes: Draw bounding boxes on a CHW image tensor

Usage

See the documentation for more details.

Slice-based API (no ndarray dependency)

All core functions have a _slice variant that operates on flat &[N] slices. This avoids coupling to a specific ndarray version (#60).

To use powerboxesrs without ndarray:

[dependencies]
powerboxesrs = { version = "0.3", default-features = false }

use powerboxesrs::iou::iou_distance_slice;

// Flat slice: [x1, y1, x2, y2, ...] with num_boxes
let boxes1 = vec![0.0, 0.0, 1.0, 1.0, 2.0, 2.0, 3.0, 3.0];
let boxes2 = vec![0.5, 0.5, 1.5, 1.5, 2.5, 2.5, 3.5, 3.5];
let iou = iou_distance_slice(&boxes1, &boxes2, 2, 2);

ndarray API (default)

With the ndarray feature (enabled by default), you get wrappers that accept ArrayView2 directly. The ndarray dependency is flexible (>=0.15, <=0.16) to minimize version conflicts.

use powerboxesrs::iou::iou_distance;
use ndarray::array;

let boxes1 = array![[0.0, 0.0, 1.0, 1.0], [2.0, 2.0, 3.0, 3.0]];
let boxes2 = array![[0.5, 0.5, 1.5, 1.5], [2.5, 2.5, 3.5, 3.5]];
let iou = iou_distance(boxes1.view(), boxes2.view());