Cascade R-CNN uses a cascade of detectors trained with progressively higher IoU thresholds to resolve overfitting and quality mismatch, achieving state-of-the-art high-quality object detection and instance segmentation on COCO and other datasets.
Rethinking ImageNet pre-training.arXiv preprint arXiv:1811.08883
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
We report competitive results on object detection and instance segmentation on the COCO dataset using standard models trained from random initialization. The results are no worse than their ImageNet pre-training counterparts even when using the hyper-parameters of the baseline system (Mask R-CNN) that were optimized for fine-tuning pre-trained models, with the sole exception of increasing the number of training iterations so the randomly initialized models may converge. Training from random initialization is surprisingly robust; our results hold even when: (i) using only 10% of the training data, (ii) for deeper and wider models, and (iii) for multiple tasks and metrics. Experiments show that ImageNet pre-training speeds up convergence early in training, but does not necessarily provide regularization or improve final target task accuracy. To push the envelope we demonstrate 50.9 AP on COCO object detection without using any external data---a result on par with the top COCO 2017 competition results that used ImageNet pre-training. These observations challenge the conventional wisdom of ImageNet pre-training for dependent tasks and we expect these discoveries will encourage people to rethink the current de facto paradigm of `pre-training and fine-tuning' in computer vision.
years
2019 4representative citing papers
T5 casts all NLP tasks as text-to-text generation, systematically explores pre-training choices, and reaches strong performance on summarization, QA, classification and other tasks via large-scale training on the Colossal Clean Crawled Corpus.
VTAB is a 19-task benchmark that measures representation quality by few-shot adaptation performance across diverse vision domains, with a controlled large-scale comparison of popular pretraining methods.
Develops a multi-task learning based adversarial training approach to improve robustness of object detectors to adversarial attacks, with experiments on PASCAL-VOC and MS-COCO.
citing papers explorer
-
Cascade R-CNN: High Quality Object Detection and Instance Segmentation
Cascade R-CNN uses a cascade of detectors trained with progressively higher IoU thresholds to resolve overfitting and quality mismatch, achieving state-of-the-art high-quality object detection and instance segmentation on COCO and other datasets.
-
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
T5 casts all NLP tasks as text-to-text generation, systematically explores pre-training choices, and reaches strong performance on summarization, QA, classification and other tasks via large-scale training on the Colossal Clean Crawled Corpus.
-
A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark
VTAB is a 19-task benchmark that measures representation quality by few-shot adaptation performance across diverse vision domains, with a controlled large-scale comparison of popular pretraining methods.
-
Towards Adversarially Robust Object Detection
Develops a multi-task learning based adversarial training approach to improve robustness of object detectors to adversarial attacks, with experiments on PASCAL-VOC and MS-COCO.