v3.1.0

What's Changed

• Fix LICENSE settings and relative links in README #358 by @dpascualhe in #359
• Upgrade streamlit and increase model file limit size by @dpascualhe in #363
• Make poetry usage consistent across actions and fix license files issue by @dpascualhe in #365
• Upgrade Pillow version to fix vulnerability by @dpascualhe in #367
• Add poetry.lock file to avoid inconsistencies by @dpascualhe in #369
• Install poetry before caching by @dpascualhe in #372
• Added Video in docs by @SakhinetiPraveena in #373
• Upgrade protobuf to fix vulnerability by @dpascualhe in #374
• Update CONTRIBUTING.md by @SakhinetiPraveena in #375
• Fix return type documentation in segmentation metrics by @tm12051990 in #381
• Bump minimatch from 3.1.2 to 3.1.5 in /docs by @dependabot[bot] in #386
• Bump flask from 3.1.2 to 3.1.3 by @dependabot[bot] in #389
• Bump werkzeug from 3.1.5 to 3.1.6 by @dependabot[bot] in #390
• fix: remove duplicate path validation block in dataset_viewer by @JaisalJain in #388
• Add reset() methods to DetectionMetricsFactory and SegmentationMetricsFactory by @hayan-gh in #384
• corrected bug for not counts gts by @iocalangiu in #395
• Improve image resize management for object detection pipeline by @dpascualhe in #400
• Fix incorrect filename in label file assertion error message by @JaisalJain in #412
• fix: handle missing or null splits in YOLODataset initialization by @karttikjangid in #399
• implement detection metrics suite and reorganize test files by @iocalangiu in #402
• [Bug-fix][UI] Truncate long file names in the datasetviewer tab by @RihaanBH-1810 in #408
• add normalized_weighted averaging method to get_averaged_metric() by @JaisalJain in #409
• [Bug-fix] prevent eval() from crashing in the case of an image with no annotations by @RihaanBH-1810 in #405
• Fix argument order bug in OnnxImageSegmentationModel initialization by @JaisalJain in #417
• Bump tornado from 6.5.4 to 6.5.5 by @dependabot[bot] in #420
• Upgrade black by @dpascualhe in #424
• Adapt detection postprocessing stage to support YOLOv26 by @dpascualhe in #425
• fix: correct COCO registry key from coco_detection to coco_image_detection by @JaisalJain in #427
• Fix YOLODataset loader when names field in data.yaml is defined as a … by @YUVAN0907 in #433
• fix: fn_count always 0 in get_overall_precision_recall_curve by @JaisalJain in #401
• [Bug-Fix][ENH] Add common func for img dir and annotation file path search using regex patterns for coco datasets. by @RihaanBH-1810 in #429
• fixes installation in windows by @redwine-1 in #435
• fix: rename duplicate 'Detection Results' heading to 'Detection Details' in inference tab by @vinitjain2005 in #446
• Remove old get_computational_cost function header by @dpascualhe in #461
• fix: write sample_pred per-sample instead of full batch pred in .tofile loop by @JaisalJain in #456
• fix: resolve silent evaluation failures by @karttikjangid in #413
• Update installation docs with PIP package info by @Saitejabojja07 in #414
• fix(evaluator): Disable Run Evaluation button when output directory is missing by @vinitjain2005 in #468
• Fix #464: Replaced harcoded paths with placeholder by @Armaan-Raisinghani in #474
• fix: save sample_pred per-sample instead of full batch pred in TF eval by @JaisalJain in #480
• fix: correct inverted missing file reporting in rellis3d and wildscenes datasets by @JaisalJain in #482
• vectorize iou computation by @Bissmella in #398
• fix: handle dict result from detection models in pm_batch by @vinitjain2005 in #488
• Fix folder selection dialog appearing twice in linux by @redwine-1 in #428
• fix: correct return type hint in build_dataset() functions by @vinitjain2005 in #484
• Bump picomatch from 2.3.1 to 2.3.2 in /docs by @dependabot[bot] in #502
• Bump requests from 2.32.5 to 2.33.0 by @dependabot[bot] in #505
• Bump pillow from 12.1.1 to 12.2.0 by @dependabot[bot] in #547
• Fix ontology LUT parameter order causing index out of bounds in image… by @sainithina in #391
• Fix: device auto-detection in GUI and specific exception handling in model loading by @Harshavardhan-Raju in #444
• docs: improve venv setup with Windows support by @priyan17singh in #472
• Fix: handle missing PyTorch dependency in inference module by @priyan17singh in #459
• docs: update installation steps for Poetry 2.x compatibility by @adenikeadewumi in #501
• Add dice score for segmentation metrics by @kayampady in #494
• Fix #521: Handle zero-area boxes in compute_iou to prevent ZeroDivisionError by @Armaan-Raisinghani in #522
• ci: add pull request trigger to CI workflow by @Armaan-Raisinghani in #528
• Bump pygments from 2.19.2 to 2.20.0 by @dependabot[bot] in #536
• Bump pytest from 8.4.2 to 9.0.3 by @dependabot[bot] in #548
• Bump gitpython from 3.1.46 to 3.1.50 by @dependabot[bot] in #574
• Bump urllib3 from 2.6.3 to 2.7.0 by @dependabot[bot] in #575
• Bump idna from 3.11 to 3.15 by @dependabot[bot] in #578
• Add nuScenes Dataset Support for 2D Object Detection by @tejasstanley in #397
• Add Cityscapes dataset adapter by @tejasstanley in #577
• Feat/cityscapes tutorial by @tejasstanley in #580
• Add support for evaluating pre-computed predictions by @dpascualhe in #582
• Bump version from 3.0.1 to 3.1.0 by @dpascualhe in #584

New Contributors

• @tm12051990 made their first contribution in #381
• @JaisalJain made their first contribution in #388
• @hayan-gh made their first contribution in #384
• @iocalangiu made their first contribution in #395
• @karttikjangid made their first contribution in #399
• @RihaanBH-1810 made their first contribution in #408
• @YUVAN0907 made their first contribution in #433
• @redwine-1 made their first contribution in #435
• @vinitjain2005 made their first contribution in #446
• @Armaan-Raisinghani made their first contribution in #474
• @Bissmella made their first contribution in #398
• @sainithina made their first contribution in #391
• @Harshavardhan-Raju made their first contribution in #444
• @priyan17singh made their first contribution in https://github.com/JdeRobot...

v3.0.1

What's Changed

• PyPI release & Github release Workflow using Poetry by @Saitejabojja07 in #355
• Fix missing diagram in website home by @dpascualhe in #356

New Contributors

• @Saitejabojja07 made their first contribution in #355

Full Changelog: v3.0.0...v3.0.1

v3.0.0

What's Changed

• Bump aiohttp from 3.8.3 to 3.12.14 in /additional_envs by @dependabot[bot] in #336
• Bump protobuf from 3.20.1 to 4.25.8 in /additional_envs by @dependabot[bot] in #335
• Bump grpcio from 1.53.0 to 1.53.2 in /additional_envs by @dependabot[bot] in #334
• Bump requests from 2.28.2 to 2.32.4 in /additional_envs by @dependabot[bot] in #337
• Bump aiohttp from 3.12.14 to 3.13.3 in /additional_envs by @dependabot[bot] in #340
• Rename DetectionMetrics -> PerceptionMetrics by @dpascualhe in #341
• Minor fixes in docs by @dpascualhe in #350
• Get rid of additional requirements by @dpascualhe in #351

Full Changelog: v2.4.0...v3.0.0

v2.4.0

What's Changed

• Add support for YOLO-styled datasets and torchscript models by @dpascualhe in #320
• Issue 323 - Changes made in preprocessing images for torch detection models. by @SakhinetiPraveena in #325
• Upgrade LiDAR segmentation by @dpascualhe in #328

Full Changelog: v2.3.0...v2.4.0

v2.3.0

What's Changed

• Fix bug in image size setting in CustomResize transform by @dpascualhe in #314
• Image detection support by @SakhinetiPraveena in #312
• Streamlit UI by @SakhinetiPraveena in #318
• Updating Docs by @SakhinetiPraveena in #319

Full Changelog: v2.2.0...v2.3.0

v2.2.0

What's Changed

🆕 Features

• Allow for an arbitrary choice of splits during evaluation by @dpascualhe in #270
• Add custom image resizing strategy for keeping aspect ratio by @dpascualhe in #272
• Allow storing resulting predictions by @dpascualhe in #274
• Add WildScenes by @dpascualhe in #276
• To support GPU acceleration using MPS in MacOS by @SakhinetiPraveena in #279
• Update computational cost estimation and add CLI support by @dpascualhe in #300

📈 Updates

• Update README.md by @jayzalani in #287
• Update pre-processing pipeline for image segmentation by @dpascualhe in #302
• Update examples by @dpascualhe in #304

🔧 Fixes

• Avoid exception when Open3D-ML and torch versions don't match by @dpascualhe in #278
• Fix typo by @jayzalani in #285
• Fix weighted average by @dpascualhe in #289

📝 Tests

• Added tests for conversion.py. by @SakhinetiPraveena in #283
• Added tests for Metrics Module by @rudrakatkar in #286

🧰 CI/CD

• Add GitHub Action for Automated Sphinx Docs Update by @meekhumor in #290
• Upgrade setuptools and wheel during build docs action by @dpascualhe in #296
• Build: complete test for remaining utils files name: io.py and lidar.py (#239) by @jayzalani in #291
• Added github actions for automatically running tests. by @SakhinetiPraveena in #294
• Added: Caching in CI/CD workflow by @jayzalani in #305

New Contributors

• @SakhinetiPraveena made their first contribution in #279
• @jayzalani made their first contribution in #285
• @rudrakatkar made their first contribution in #286
• @meekhumor made their first contribution in #290

Full Changelog: v2.1.3...v.2.1.4

v2.1.3

What's Changed

• 261 render lidar point clouds as images by @dpascualhe in #262
• Add functions to calculate Precision, Recall and F1score. by @Afreen-Kazi-1 in #255
• Add support for RUGD dataset by @dpascualhe in #264
• Update docs by @dpascualhe in #266
• Update metrics & improvements in datasets by @dpascualhe in #267

New Contributors

• @Afreen-Kazi-1 made their first contribution in #255

Full Changelog: v2.1.2...v2.1.3

v2.1.2

What's Changed

• Fixes #252- Add a How to Contribute Section by @vm7151810 in #250
• Add confusion matrix by @dpascualhe in #253
• Add image segmentation tutorial by @dpascualhe in #254
• Add support for native tensorflow and pytorch models by @dpascualhe in #260
• Computational cost estimation by @dpascualhe in #258

New Contributors

• @vm7151810 made their first contribution in #250

Full Changelog: v2.1.1...v2.1.2

v2.1.1

What's Changed

• Bump glob-parent and onchange in /docs by @dependabot in #227
• Add CLI by @dpascualhe in #249
• Increase version by @dpascualhe in #251

Full Changelog: v2.1.0...v2.1.1

v2.1.0

What's Changed

• Add support for LiDAR segmentation by @dpascualhe in #246. More specifically, added support for GOOSE and Rellis3D LiDAR segmentation datasets and PyTorch models (validated for RandLA-Net and KPConv from Open3D-ML.
• Update docs and increase version by @dpascualhe in #248

Full Changelog: v2.0.0...v2.1.0