Ben Lonnqvist, Harshitha Machiraju, Michael H. Herzog
In a recent article, Guo et al. [arXiv:2206.11228] report that adversarially trained neural representations in deep networks may already be as robust as corresponding primate IT neural representations. While we find the paper's primary experiment illuminating, we have doubts about the interpretation and phrasing of the results presented in the paper.
Ben Lonnqvist, Zhengqing Wu, Michael H. Herzog
Humans are able to segment images effortlessly without supervision using perceptual grouping. Here, we propose a counter-intuitive computational approach to solving unsupervised perceptual grouping and segmentation: that they arise because of neural noise, rather than in spite of it. We (1) mathematically demonstrate that under realistic assumptions, neural noise can be used to separate objects from each other; (2) that adding noise in a DNN enables the network to segment images even though it was never trained on any segmentation labels; and (3) that segmenting objects using noise results in segmentation performance that aligns with the perceptual grouping phenomena observed in humans, and is sample-efficient. We introduce the Good Gestalt (GG) datasets -- six datasets designed to specifically test perceptual grouping, and show that our DNN models reproduce many important phenomena in human perception, such as illusory contours, closure, continuity, proximity, and occlusion. Finally, we (4) show that our model improves performance on our GG datasets compared to other tested unsupervised models by $24.9\%$. Together, our results suggest a novel unsupervised segmentation method requiring few assumptions, a new explanation for the formation of perceptual grouping, and a novel potential benefit of neural noise.
Ben Lonnqvist, Elsa Scialom, Abdulkadir Gokce et al.
Despite the tremendous success of deep learning in computer vision, models still fall behind humans in generalizing to new input distributions. Existing benchmarks do not investigate the specific failure points of models by analyzing performance under many controlled conditions. Our study systematically dissects where and why models struggle with contour integration -- a hallmark of human vision -- by designing an experiment that tests object recognition under various levels of object fragmentation. Humans (n=50) perform at high accuracy, even with few object contours present. This is in contrast to models which exhibit substantially lower sensitivity to increasing object contours, with most of the over 1,000 models we tested barely performing above chance. Only at very large scales ($\sim5B$ training dataset size) do models begin to approach human performance. Importantly, humans exhibit an integration bias -- a preference towards recognizing objects made up of directional fragments over directionless fragments. We find that not only do models that share this property perform better at our task, but that this bias also increases with model training dataset size, and training models to exhibit contour integration leads to high shape bias. Taken together, our results suggest that contour integration is a hallmark of object vision that underlies object recognition performance, and may be a mechanism learned from data at scale.
Johannes Mehrer, Ben Lonnqvist, Anna Mitola et al.
Brain stimulation is a powerful tool for understanding cortical function and holds promise for therapeutic interventions in neuropsychiatric disorders. Initial visual prosthetics apply electric microstimulation to early visual cortex which can evoke percepts of simple symbols such as letters. However, these approaches are fundamentally limited by hardware constraints and the low-level representational properties of this cortical region. In contrast, higher-level visual areas encode more complex object representations and therefore constitute a promising target for stimulation - but determining representational targets that reliably evoke object-level percepts constitutes a major challenge. We here introduce a computational framework to causally model and guide stimulation of high-level cortex, comprising three key components: (1) a perturbation module that translates microstimulation parameters into spatial changes to neural activity, (2) topographic models that capture the spatial organization of cortical neurons and thus enable prototyping of stimulation experiments, and (3) a mapping procedure that links model-optimized stimulation sites back to primate cortex. Applying this framework in two macaque monkeys performing a visual recognition task, model-predicted stimulation experiments produced significant in-vivo changes in perceptual choices. Per-site model predictions and monkey behavior were strongly correlated, underscoring the promise of model-guided stimulation. Image generation further revealed a qualitative similarity between in-silico stimulation of face-selective sites and a patient's report of facephenes. This proof-of-principle establishes a foundation for model-guided microstimulation and points toward next-generation visual prosthetics capable of inducing more complex visual experiences.
Ben Lonnqvist, Alasdair D. F. Clarke, Ramakrishna Chakravarthi
Object recognition is a primary function of the human visual system. It has recently been claimed that the highly successful ability to recognise objects in a set of emergent computer vision systems---Deep Convolutional Neural Networks (DCNNs)---can form a useful guide to recognition in humans. To test this assertion, we systematically evaluated visual crowding, a dramatic breakdown of recognition in clutter, in DCNNs and compared their performance to extant research in humans. We examined crowding in three architectures of DCNNs with the same methodology as that used among humans. We manipulated multiple stimulus factors including inter-letter spacing, letter colour, size, and flanker location to assess the extent and shape of crowding in DCNNs. We found that crowding followed a predictable pattern across architectures that was different from that in humans. Some characteristic hallmarks of human crowding, such as invariance to size, the effect of target-flanker similarity, and confusions between target and flanker identities, were completely missing, minimised or even reversed. These data show that DCNNs, while proficient in object recognition, likely achieve this competence through a set of mechanisms that are distinct from those in humans. They are not necessarily equivalent models of human or primate object recognition and caution must be exercised when inferring mechanisms derived from their operation.