Ashish Gupta
I am a multi-disciplinary scientist and software developer, researching the technology domains of machine and deep learning, computer vision, data science, signal and image processing.
I am a professional member of the IEEE, ACM, AAAI and review papers for the journals CVIU, IEEE-TIP, PE&RS and several technical conferences.
My driving vision is designing thinking machines that function symbiotically with high dimensional and dynamic data in complex systems.
I wear several hats including scientist, engineer, teacher, and entrepreneur. Besides co-founding Ubihere, an intelligent geo-spatial localization solutions company, I serve as technical adviser and collaborator on multiple technology ventures. In academia, I mentor graduate students and also collaborate on research grant proposals to funding organizations including NSF, NGA, NASA, NOAA, and DoD.
My research focus at present is development of Deep Learning methods, especially Adversarial Networks towards media forensic analysis on a DARPA funded project. I have ongoing collaborations in the indoor navigation and automated driving technology space.
I look forward to opportunities to work with technologists in academia, industry or start-up space.
Supervisors: Alper Yilmaz, Frederic Jurie, Josef Kittler, K S Venkatesh, Christopher Kanan, Chistye Sisson, Pingbo Tang, and Marc Kozak
Address: 76 (CAR) 2214, 54 Lombard Memorial Dr., Rochester, NY 14623
I am a professional member of the IEEE, ACM, AAAI and review papers for the journals CVIU, IEEE-TIP, PE&RS and several technical conferences.
My driving vision is designing thinking machines that function symbiotically with high dimensional and dynamic data in complex systems.
I wear several hats including scientist, engineer, teacher, and entrepreneur. Besides co-founding Ubihere, an intelligent geo-spatial localization solutions company, I serve as technical adviser and collaborator on multiple technology ventures. In academia, I mentor graduate students and also collaborate on research grant proposals to funding organizations including NSF, NGA, NASA, NOAA, and DoD.
My research focus at present is development of Deep Learning methods, especially Adversarial Networks towards media forensic analysis on a DARPA funded project. I have ongoing collaborations in the indoor navigation and automated driving technology space.
I look forward to opportunities to work with technologists in academia, industry or start-up space.
Supervisors: Alper Yilmaz, Frederic Jurie, Josef Kittler, K S Venkatesh, Christopher Kanan, Chistye Sisson, Pingbo Tang, and Marc Kozak
Address: 76 (CAR) 2214, 54 Lombard Memorial Dr., Rochester, NY 14623
less
InterestsView All (11)
Uploads
Papers by Ashish Gupta
"
meaningful dictionary, with regards to correspondence between object parts and multiple sub-manifolds, and is not intended to compete with state-of-the-art methods like sparse coding. It is specially pertinent for the future for learning a dictionary with increasing complexity of visual categories.
Books by Ashish Gupta
Using Computer Vision towards geospatial localization in GPS-denied or degraded environment.
Several significant algorithms were tested for each stage of the synthesis procedure to find the techniques optimal for static handwriting data: entropy based threshold for character image extraction; Kuwahara filter for de-noising; Zhang-Suen algorithm for skeletonization; distance transform for control-point selection; shape-context descriptor for control-point correspondence search; thin plate splines for control-point transformation; and interpolating splines for generating stroke curves. Empirical results indicate that the novel combination of handwriting specific algorithms in this thesis can generate realistic synthetic handwriting in a given writer’s unique style, of satisfactory quality.
Towards incorporating this structure in the learning algorithms, this thesis analyses two facets of feature data to discover relevant structure. The first is structure amongst the sub-spaces of the feature descriptor. Several sub-space embedding techniques that use global or local information to compute a projection function are analysed. A novel entropy based measure of structure in the embedded descriptors suggests that relevant structure has local extent. The second is structure amongst the partitions of feature space. Hard partitioning of feature space leads to ambiguity in feature encoding. To address this issue, novel fuzzy logic based dictionary learning and feature encoding algorithms are employed that are able to model the local feature vectors distributions and provide performance benefits.
To estimate structure amongst sub-spaces, co-clustering is used with a training descriptor data matrix to compute groups of sub-spaces. A dictionary learnt on feature vectors embedded in these multiple sub-manifolds is demonstrated to model data better than a dictionary learnt on feature vectors embedded in a single sub-manifold computed using principal components. In a similar manner, co-clustering is used with encoded feature data matrix to compute groups of dictionary elements - referred to as `topics'. A topic dictionary is demonstrated to perform better than a regular dictionary of comparable size. Both these results suggest that the groups of sub-spaces and dictionary elements have semantic relevance.
All the methods developed here have been viewed from the unifying perspective of matrix factorization, where a data matrix is decomposed to two factor matrices which are interpreted as a dictionary matrix and a co-efficient matrix. Sparse coding methods, which are currently enjoying much success, can be viewed as matrix factorization with a regularization constraint on the vectors of the dictionary or co-efficient matrices. With regards to sub-space embedding, the sparse principal component analysis is one such method that induces sparsity amongst the sub-spaces selected to represent each descriptor. Similarly, Lasso is used to induce sparsity in feature encoding by using only a sub-set of dictionary elements to represent each image. While these methods are effective, they disregard structure in the data matrix. To improve on this, structured sparse principal component analysis is used in conjunction with co-clustered groups of sub-spaces to induce sparsity at group level. The resultant structured sparse sub-
manifold dictionary is demonstrated to provide performance benefits. In a similar manner, group Lasso is used with co-clustered groups of dictionary elements to induce sparsity in terms of topics. The structured sparse encoding is demonstrated to improve aggregate performance in comparison to a regular sparse coding.
In conclusion, this thesis estimates structure in descriptor sub-spaces and learnt dictionary, uses co-clustering to compute semantically relevant sub-manifolds and topic dictionary, and finally incorporates the estimated structure in sparse coding methods, demonstrating performance gain for visual category recognition.
Talks by Ashish Gupta