Abstract: The disclosure provides computer-implemented method for training a vision-language machine learning, ML, model to classify images depicting novel or known classes. The method comprises obtaining a first training dataset comprising a plurality of class names and training the vision-language ML model. The training method comprises: generating at least one augmented textual prompt; inputting the at least one augmented textual prompt into a frozen text encoder; outputting a first text embedding for each augmented textual prompt; generating a plurality of first inputs by concatenating each learnable soft prompt from a plurality of learnable soft prompts; inputting the class names and the plurality of first inputs into the frozen text encoder; outputting a second text embedding for each first input; and minimizing a cross-entropy text-to-text loss between the first text embeddings and the second text embeddings.

Abstract: A method of controlling an apparatus for performing video action classification using a trained machine learning, ML, model, the method includes receiving a plurality of frames of a video, inputting, into the trained ML model, the plurality of frames, identifying an actor in the plurality of frames, wherein the actor performs an action in the plurality of frames, and based on the actor being identified, classifying the action performed by the actor.

Abstract: Broadly speaking, the present techniques generally relate to a computer-implemented method for analysing images or videos using a machine learning, ML, model and recognising actions within the image or video. Advantageously, the present techniques provide a ML model which is of a size suitable for implementation on constrained resource devices, such as smartphones. Furthermore, the present techniques provide a ML model which is more computationally efficient (from a processor and memory perspective), without any loss in accuracy.

Abstract: Broadly speaking, the present techniques generally relate to a method and apparatus for video recognition, and in particular relate to a computer-implemented method for performing video recognition using a transformer-based machine learning, ML, model. Put another way, the present techniques provide new methods of image processing in order to automatically extract feature information from a video.

Abstract: Broadly speaking, the present techniques generally relate to machine learning models comprising neural network layers, in which the quantisation level of each layer of the model can be independently selected at run-time. In particular, the present application relates to a computer-implemented method for analysing input data on a device using a trained machine learning, ML, model, comprising independently selecting a quantisation level for each of a plurality of network layers of the model at runtime. The present application also relates to a computer-implemented method of training a machine learning model so that the quantisation level of each of the plurality of network layers is independently selectable at runtime. A single trained model with a single set of weights can therefore be deployed, with the quantisation of each layer selected at runtime to suit the capabilities of the device and available resource.