| Year | Title | Author | Publication | Code | Tasks | Notes | Datasets | Notions |
|---|---|---|---|---|---|---|---|---|
| 2021 | Semi-Supervised Active Learning With Temporal Output Discrepancy | Huang et al. | ICCV | code | Temporal Output Discrepancy | Informative, ResNet-18,None, PT+FT, Hard |
Cifar-10, Cifar-100, SVHN, and Caltech-101. | |
| 2021 | Semi-Supervised Active Learning for Semi-Supervised Models: Exploit Adversarial Examples With Graph-Based Virtual Labels | Guo et al. | ICCV | - | Image Classification | diversity, Graph-based+ResNet, Adversatial,PT+FT, Hard |
CIFAR- 10 [22] and CIFAR-100 | |
| 2021 | Influence Selection for Active Learning | Liu et al. | ICCV | - | Image Classification and Object detection | expected gradient, Any Neural Networks, None, Tra, Hard |
CIFAR10, VOC2012, COCO | |
| 2021 | Contrastive Coding for Active Learning Under Class Distribution Mismatch | Du et al. | ICCV | code | semantic and distinctive, ResNet18, contrastive learning, PT+FT, Hard |
CIFAR10, CIFAR100, artificial cross-dataset | Class Distribution Mismatch | |
| 2021 | ReDAL: Region-Based and Diversity-Aware Active Learning for Point Cloud Semantic Segmentation | Wu et al. | ICCV | - | Point Cloud Semantic Segmentation | Diversity, MinkowskiNet/SPVCNN, None, PT+FT, Hard |
S3DIS and Se- manticKITTI datasets | |
| 2021 | Active Learning for Deep Object Detection via Probabilistic Modeling | Choi et al. | ICCV | code | object detection | epistemic uncertainty, GMM-based, None, PT+FT, Hard |
PASCAL, VOC, MS COCO | |
| 2021 | Active Learning for Lane Detection: A Knowledge Distillation Approach | Peng et al. | ICCV | - | uncer- tainty and diversity metrics, PointLaneNet/UFLD, Knowledge Distillation, PT+FT, Hard |
CULane [29] and LLAMAS [4] | knowledge distillation approach |