Trame-annotations

Fix

* fix(BoxAnnotations): limit canvas width+height to display size ([`fffb4da`](https://github.com/Kitware/trame-annotations/commit/fffb4daefed9a672540720bd1d2751adbf4614da))

Fix

* fix(hatch): improve hook code ([`4cb468c`](https://github.com/Kitware/trame-annotations/commit/4cb468c2f3ce90f192c93f9515025fe9e848609a))

* fix(hatch): windows build path handling ([`e4d6603`](https://github.com/Kitware/trame-annotations/commit/e4d660349a4d0dfc6ba8ace5eef1c022d4f7cf9f))

Fix

* fix(hatch): windows build path handling ([`7a474e0`](https://github.com/Kitware/trame-annotations/commit/7a474e0beb2fb72fc0ddcbe3c796a3361d873016))

Feature

* feat(ClassificationAnnotations): stack multiple dots ([`4cabc19`](https://github.com/Kitware/trame-annotations/commit/4cabc1997d06545baa7ae87b5909a0a378d8bf03))

* feat(ImageDetection): show score as percentage ([`ba2aad1`](https://github.com/Kitware/trame-annotations/commit/ba2aad1fbeed5b262e9beeb8411640d765f1e2aa))

* feat(ImageDetection): red reserved for missing category annotations ([`23d812e`](https://github.com/Kitware/trame-annotations/commit/23d812e77b4469cf010837f2c8ea0ab87955ee01))

* feat(ImageDetection): add score threshold prop

Filters annotations. ([`e229398`](https://github.com/Kitware/trame-annotations/commit/e2293988c2c33bd1804c1514dc2024854490a869))

* feat(ImageDetection): add 4 more category colors ([`e1465db`](https://github.com/Kitware/trame-annotations/commit/e1465dbeafc3a617a249d8d0e8ba409e36d1c476))

* feat(ImageDetection): show score for annotations ([`4eccedf`](https://github.com/Kitware/trame-annotations/commit/4eccedfb0fe33cc2841c41564ae62aaea76a1a6c))

* feat(ImageDetection): hover over dot to see classes ([`867f415`](https://github.com/Kitware/trame-annotations/commit/867f415c2edc5fb59aec01d8660ae99f6e82e8ba))

Fix

* fix(ImageDetection): emit hover event in mouseleave function

There is an error in Javascript that borks client side trame code
when calling emit(&39;hover&39;) in a watchEffect. Steps
1. In nrtk-explorer
2. Turn on Model Inference
3. Switch to Grid view for image list
4. Error in JS console:
NotReadableError: The requested file could not be read, typically due to permission problems that have occurred after a reference to a file was acquired.
in
t.onmessage = async u => {
const p = await c.processChunk(u.data, h);
p && l(p)
} ([`d38cd84`](https://github.com/Kitware/trame-annotations/commit/d38cd84c7f4777cd43b4e6ef70221d2f5ccb90c4))

Refactor

* refactor(ImageDetection): extract BoxAnnotations component ([`784aedb`](https://github.com/Kitware/trame-annotations/commit/784aedb4f986bfd57b0bff61f1eb4c66d4dfa4ba))

* refactor(ImageDetection): extract ClassificationAnnotations component ([`f974fcf`](https://github.com/Kitware/trame-annotations/commit/f974fcfe15b99a3e3e13abd4560f7a311b54b621))

* refactor(ImageDetection): extract AnnotationPopup component ([`49d6871`](https://github.com/Kitware/trame-annotations/commit/49d6871cc21026979903525e57570e94537f910d))

* refactor(ImageDetection): declarative style for mouse events ([`8faee9a`](https://github.com/Kitware/trame-annotations/commit/8faee9a2690bee984853a32e383bc7b728464ad4))

Unknown

* Merge pull request 7 from Kitware/hover-classes

Hover over dot to show image classifications ([`f18eb65`](https://github.com/Kitware/trame-annotations/commit/f18eb656496ce862122123d2f0ca38937a325627))

Feature

* feat(ImageDetection): add whole image classification support ([`3177d19`](https://github.com/Kitware/trame-annotations/commit/3177d19327947fd8f91927bbec14767ce1f4236a))

Unknown

* Merge pull request 5 from Kitware/classes

feat(ImageDetection): add whole image classification support ([`0ba98f3`](https://github.com/Kitware/trame-annotations/commit/0ba98f33515ba4dcd0205a95740d385430b0b25e))

Feature

* feat(ImageDetection): additive blend mode for annotations ([`873accc`](https://github.com/Kitware/trame-annotations/commit/873accc072760681dfb9ad9ae4fcce9e980dc25e))

* feat(ImageDetection): add line_width and line_opacity props ([`ea6aaa3`](https://github.com/Kitware/trame-annotations/commit/ea6aaa303f388a7e461d3c0aab7dcf1ac9dafaa2))

Fix

* fix(ImageDetection): fallback to line width default correctly ([`cb3c237`](https://github.com/Kitware/trame-annotations/commit/cb3c237529740a420ecb97046ef5a6ec750c834e))

* fix(ImageDetection): set bbox line width in display pixels

If an image was large, but displayed at a smaller size,
the bounding box line was very thin and hard to see. ([`fd74d92`](https://github.com/Kitware/trame-annotations/commit/fd74d92c5a6c88725de37708b9a2f5a702e59731))

Unknown

* Merge pull request 4 from Kitware/line-width

Box line width in display pixels ([`71f4d5a`](https://github.com/Kitware/trame-annotations/commit/71f4d5aa8640f580c1a1399bb8d9c6f7e1bf34d8))