We introduce Patch Aligned Contrastive Learning (PACL), a modified compatibility function for CLIP's contrastive loss, intending to train an alignment between the patch tokens of the vision encoder and the CLS token of the text encoder.
In this paper, we define 2 categories of OoD data using the subtly different concepts of perceptual/visual and semantic similarity to in-distribution (iD) data.
We extend Deep Deterministic Uncertainty (DDU), a method for uncertainty estimation using feature space densities, to semantic segmentation.
Reliable uncertainty from deterministic single-forward pass models is sought after because conventional methods of uncertainty quantification are computationally expensive.
We study batch normalisation in the context of variational inference methods in Bayesian neural networks, such as mean-field or MC Dropout.
To facilitate the use of focal loss in practice, we also provide a principled approach to automatically select the hyperparameter involved in the loss function.
When combined with temperature scaling, focal loss, whilst preserving accuracy and yielding state-of-the-art calibrated models, also preserves the confidence of the model's correct predictions, which is extremely desirable for downstream tasks.
Many applications require a camera to be relocalised online, without expensive offline training on the target scene.
Deep learning has been revolutionary for computer vision and semantic segmentation in particular, with Bayesian Deep Learning (BDL) used to obtain uncertainty maps from deep models when predicting semantic classes.
Ranked #6 on Anomaly Detection on Fishyscapes