What's Changed
* Update unit test and golden numbers in KPL by qlzh727 in https://github.com/keras-team/keras-cv/pull/864
* Support no image size in PascalVOC loader by LukeWood in https://github.com/keras-team/keras-cv/pull/862
* Make separate benchmark methods for single and multi-gpu benchmarks by ianstenbit in https://github.com/keras-team/keras-cv/pull/867
* revert keypoint refactoring. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/869
* Add deeplab_v3 model with keras cv components only by qlzh727 in https://github.com/keras-team/keras-cv/pull/856
* Update the instructions to update the cloudbuild cluster by LukeWood in https://github.com/keras-team/keras-cv/pull/832
* Address review comments from 874 by qlzh727 in https://github.com/keras-team/keras-cv/pull/875
* Support mixed-precision training in basic_training script by ianstenbit in https://github.com/keras-team/keras-cv/pull/878
* Add FasterRCNN model. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/870
* Fix the unit test for segmentation head serialization. by qlzh727 in https://github.com/keras-team/keras-cv/pull/882
* Add Atrous Spatial Pyramid Pooling as a feature decoder layer by qlzh727 in https://github.com/keras-team/keras-cv/pull/884
* Add contributing docs about new model weights by ianstenbit in https://github.com/keras-team/keras-cv/pull/889
* Add devcontainer CI by bhack in https://github.com/keras-team/keras-cv/pull/893
* Improve FasterRCNN model quality. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/898
* Minor performance improvements to basic ImageNet training by ianstenbit in https://github.com/keras-team/keras-cv/pull/902
* Add pre-trained weights for EfficientNetV2B0 by ianstenbit in https://github.com/keras-team/keras-cv/pull/906
* Filter pull request directory by bhack in https://github.com/keras-team/keras-cv/pull/899
* Make RandomFlip perform horizontal flipping only by default by ianstenbit in https://github.com/keras-team/keras-cv/pull/903
* Add distributed training, and further improve training quality. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/914
* Add some pre-trained weight info to README by ianstenbit in https://github.com/keras-team/keras-cv/pull/913
* Adds the VAEEncoder to StableDiffusion by LukeWood in https://github.com/keras-team/keras-cv/pull/873
* Allow compute_iou to take boxes of different shape. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/919
* Add pretrained EfficientNetV2b1 weights by ianstenbit in https://github.com/keras-team/keras-cv/pull/923
* update readme and roadmap. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/934
* Fix typo in name of devcontainer workflow by ianstenbit in https://github.com/keras-team/keras-cv/pull/940
* Add model fit for FasterRCNN. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/925
* Update MobileNet docs to point to the correct paper by ianstenbit in https://github.com/keras-team/keras-cv/pull/948
* Add pascal voc 2012 dataset with segmentation data. by qlzh727 in https://github.com/keras-team/keras-cv/pull/943
* Add RandomCropAndResize to ImageNet classification training script by ianstenbit in https://github.com/keras-team/keras-cv/pull/935
* Add better weights for ResNet50V2 by ianstenbit in https://github.com/keras-team/keras-cv/pull/952
* Fixed efficientnet v1 correct pad. by sebastian-sz in https://github.com/keras-team/keras-cv/pull/927
* Update the RetinaNet example by LukeWood in https://github.com/keras-team/keras-cv/pull/942
* Add tf.function(jit_compile=True) for the PNG decode function. by qlzh727 in https://github.com/keras-team/keras-cv/pull/953
* Add IoU3D as a custom c++ op (CPU) by ianstenbit in https://github.com/keras-team/keras-cv/pull/890
* bug fix in image_encoder.py by LukeWood in https://github.com/keras-team/keras-cv/pull/955
* Pre-trained weights for EfficientNetV2B2 by ianstenbit in https://github.com/keras-team/keras-cv/pull/956
* Training script improvement proposal by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/908
* Devcontainer GPU support 945 by jacoverster in https://github.com/keras-team/keras-cv/pull/946
* Fix build target for custom ops by ianstenbit in https://github.com/keras-team/keras-cv/pull/959
* Step 1/3. add dummy semantic segmentation training script. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/957
* Step 2/3: add Sbd data by tanzhenyu in https://github.com/keras-team/keras-cv/pull/961
* Fix casting issue with basic training learning rate schedule by ianstenbit in https://github.com/keras-team/keras-cv/pull/965
* Add waymo dataset loader by ianstenbit in https://github.com/keras-team/keras-cv/pull/966
* Add weights for EfficientNetV2S by ianstenbit in https://github.com/keras-team/keras-cv/pull/971
* citing reference. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/972
* Fix reproducability issue of seeded StableDiffusion runs by ianstenbit in https://github.com/keras-team/keras-cv/pull/962
* Added mixed_precision support for preprocessing layers. by sebastian-sz in https://github.com/keras-team/keras-cv/pull/967
* Save best weights only for basic_training by ianstenbit in https://github.com/keras-team/keras-cv/pull/977
* 630/add efficientnet lite by sebastian-sz in https://github.com/keras-team/keras-cv/pull/921
* Shuffle dataset every epoch for basic_training by ianstenbit in https://github.com/keras-team/keras-cv/pull/983
* Track encoded boxes in the RetinaNet in a Keras metric by LukeWood in https://github.com/keras-team/keras-cv/pull/964
* Step 3/3: Update DeeplabV3 model by tanzhenyu in https://github.com/keras-team/keras-cv/pull/975
* Add README about Waymo Open Dataset dependency by ianstenbit in https://github.com/keras-team/keras-cv/pull/985
* Lazy-load components of StableDiffusion and make them visible to users by ianstenbit in https://github.com/keras-team/keras-cv/pull/989
* Add text-conditioned in-painting functionality to Keras Stable Diffusion by bobqywei in https://github.com/keras-team/keras-cv/pull/973
* Add support for `add_tokens()` to StableDiffusion in order to support TextualInversion by LukeWood in https://github.com/keras-team/keras-cv/pull/831
* Add bounding box support to resizing by LukeWood in https://github.com/keras-team/keras-cv/pull/861
* Implement bounding boxes for RandomCropAndResize by AmedeoBiolatti in https://github.com/keras-team/keras-cv/pull/994
* Update README.md with disclaimer by LukeWood in https://github.com/keras-team/keras-cv/pull/993
* Avoid retracing with python lazy_loading by bhack in https://github.com/keras-team/keras-cv/pull/997
* Restructure StableDiffusion namespace and folder hierarchy by ianstenbit in https://github.com/keras-team/keras-cv/pull/1002
* fix serializationt test. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1004
* SimCLR Training Script by ianstenbit in https://github.com/keras-team/keras-cv/pull/717
* Add `seed=seed`, param_name to factor parsing for layers.Mosaic() by LukeWood in https://github.com/keras-team/keras-cv/pull/881
* Adding ViT Components/Layers by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/929
* Add base augmentation layer for 3D preception. by lengzq in https://github.com/keras-team/keras-cv/pull/986
* [Object Detection] Update retinanet predict loop to enable TPU support by ianstenbit in https://github.com/keras-team/keras-cv/pull/1017
* [Object Detection] Introduce a callback-based coco metric evaluator by ianstenbit in https://github.com/keras-team/keras-cv/pull/1000
* fixed transformer encoder dims per head by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/1013
* Revert "[Object Detection] Update retinanet predict loop to enable TPU support" by ianstenbit in https://github.com/keras-team/keras-cv/pull/1022
* ResNet and ResNetV2 - Typo fix in docstrings by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/1023
* Adding RegNets to keras_cv by AdityaKane2001 in https://github.com/keras-team/keras-cv/pull/739
* Add global_z_rotation data augmentation and wrap_angle_rad helper function. by lengzq in https://github.com/keras-team/keras-cv/pull/1012
* [Backbones] Add pre-trained weights for DarkNet53 by ianstenbit in https://github.com/keras-team/keras-cv/pull/1029
* [Object Detection] RetinaNet: custom predict_function by ianstenbit in https://github.com/keras-team/keras-cv/pull/1021
* [Object Detection] Speed up the PyCOCO callback by only calling model.predict once by ianstenbit in https://github.com/keras-team/keras-cv/pull/1026
* Add mixed precision for deeplab. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1032
* Update group_normalization.py by charbull in https://github.com/keras-team/keras-cv/pull/1035
* [Backbone Training] added 50/50 between cutmix and mixup, changed default hold steps by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/980
* IoULoss documentation typo fix by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/1038
* Add four global data augmentations for 3D by lengzq in https://github.com/keras-team/keras-cv/pull/1028
* Add pre-trained weights for CSPDarkNet by ianstenbit in https://github.com/keras-team/keras-cv/pull/1036
* [Draft] Polish the RetinaNet by LukeWood in https://github.com/keras-team/keras-cv/pull/984
* fix for dynamic shape. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1042
* fix fasterrcnn train dataset. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1046
* [Stable Diffusion] Noise scheduler for fine-tuning and textual inversion by ianstenbit in https://github.com/keras-team/keras-cv/pull/1045
* polish roadmap and readme. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1009
* Introduced a new internal util to preserve relative format by quantumalaviya in https://github.com/keras-team/keras-cv/pull/1016
* Add group_points_by_bounding_boxes data augmentation. by lengzq in https://github.com/keras-team/keras-cv/pull/1040
* [Semantic Segmentation] Add first segmentation weights (credit tanzhenyu) by ianstenbit in https://github.com/keras-team/keras-cv/pull/1059
* (Ragged OD Pipeline Step 1/3) Support RaggedImages and Bounding Boxes in BaseImageAugmentationLayer by LukeWood in https://github.com/keras-team/keras-cv/pull/1047
* [Object Detection] Remove train_time_metrics from RetinaNet by ianstenbit in https://github.com/keras-team/keras-cv/pull/1041
* [3D Object Detection] Make iou_3d a function by ianstenbit in https://github.com/keras-team/keras-cv/pull/1060
* Step 1/4 [ObjectDetection]: allow PyCOCOCallback to take apply_nms by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1061
* Implement bounding boxes for random choice by AmedeoBiolatti in https://github.com/keras-team/keras-cv/pull/998
* Revert "Implement bounding boxes for random choice" by ianstenbit in https://github.com/keras-team/keras-cv/pull/1065
* (Step 2/3) creates a JitteredResize layer by LukeWood in https://github.com/keras-team/keras-cv/pull/1053
* Update stable_diffusion `num_steps` to be 50 by LukeWood in https://github.com/keras-team/keras-cv/pull/1067
* fix several failing test / scripts due to recent PR merge. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1069
* Introducing randomly pasting objects data augmentation. by lengzq in https://github.com/keras-team/keras-cv/pull/1058
* Re-introduce evaluate_train_step_metrics API by LukeWood in https://github.com/keras-team/keras-cv/pull/1071
* Switch to one-configuration-per-architecture GCB testing for backbones by ianstenbit in https://github.com/keras-team/keras-cv/pull/1070
* move fasterrcnn to a new path. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1072
* Fix patch that somehow got dropped by LukeWood in https://github.com/keras-team/keras-cv/pull/1076
* Add data transformer for Waymo Open Dataset by Guowang in https://github.com/keras-team/keras-cv/pull/1066
* Introduce swap background data augmentation. by lengzq in https://github.com/keras-team/keras-cv/pull/1078
* 3D data augmentation cleanup by lengzq in https://github.com/keras-team/keras-cv/pull/1082
* [3D Object Detection] Waymo dataloading + augmentation flow example by ianstenbit in https://github.com/keras-team/keras-cv/pull/1083
* Customized op for WithinBox3d. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1020
* Introduce random drop box 3D data augmentation. by lengzq in https://github.com/keras-team/keras-cv/pull/1084
* Add a note about mixed precision to SD `inpaint` by ianstenbit in https://github.com/keras-team/keras-cv/pull/1088
* replace is_within_box3d with is_within_any_box3d. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1087
* Group points by boxes v2 by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1090
* [3D Object Detection] Add initial 3D box format by ianstenbit in https://github.com/keras-team/keras-cv/pull/1064
* Fix ragged_tensor resize new case by bhack in https://github.com/keras-team/keras-cv/pull/891
* Fix broken build by ianstenbit in https://github.com/keras-team/keras-cv/pull/1093
* Introduce frustum based drop points by lengzq in https://github.com/keras-team/keras-cv/pull/1091
* Fix typos in tests and add 3d augmentations to init. by lengzq in https://github.com/keras-team/keras-cv/pull/1096
* Gate RandomCopyPasteTest on TEST_CUSTOM_OPS flag by ianstenbit in https://github.com/keras-team/keras-cv/pull/1097
* [Mixed Precision Support] MixUp Layer by DavidLandup0 in https://github.com/keras-team/keras-cv/pull/1092
* Fixed a Minor Typo by aflah02 in https://github.com/keras-team/keras-cv/pull/1094
* Fix RetinaNet training on TPU. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1113
* Prefix custom op names with kcv by ianstenbit in https://github.com/keras-team/keras-cv/pull/1115
* Introduce frustum-based point feature noise. by lengzq in https://github.com/keras-team/keras-cv/pull/1098
* Delete .so files by ianstenbit in https://github.com/keras-team/keras-cv/pull/1116
* transform records. by tanzhenyu in https://github.com/keras-team/keras-cv/pull/1089
* (OD input pipeline Step 3/4) Add RandomAspectRatio layer by LukeWood in https://github.com/keras-team/keras-cv/pull/1054
New Contributors
* DavidLandup0 made their first contribution in https://github.com/keras-team/keras-cv/pull/908
* jacoverster made their first contribution in https://github.com/keras-team/keras-cv/pull/946
* bobqywei made their first contribution in https://github.com/keras-team/keras-cv/pull/973
* AmedeoBiolatti made their first contribution in https://github.com/keras-team/keras-cv/pull/994
* lengzq made their first contribution in https://github.com/keras-team/keras-cv/pull/986
* charbull made their first contribution in https://github.com/keras-team/keras-cv/pull/1035
* Guowang made their first contribution in https://github.com/keras-team/keras-cv/pull/1066
* aflah02 made their first contribution in https://github.com/keras-team/keras-cv/pull/1094
**Full Changelog**: https://github.com/keras-team/keras-cv/compare/v0.3.4...v0.3.5