With the advancement in information and communication technologies, sensing devices have now become pervasive. The pervasiveness of camera devices has enabled recording of video data at anytime and anywhere. It gives rise to a massive amount of untrimmed video data being produced, which consist of several human-related activities and actions including some background activities as well. It is important to detect the actions of interest in such long and untrimmed videos so that it can be further used in numerous applications i.e., video analysis, video summarization, surveillance, retrieval and captioning etc. This thesis targets temporal human action detection in long and untrimmed videos. Given a long and untrimmed video, the task of the temporal action detection is to detect starting and ending time of all occurrences of actions of interest and to predict action label of the detected intervals. Detecting human actions in long untrimmed videos is important but a challenging problem because of the unconstrained nature of long untrimmed videos in both space and time. In this work we solve the temporal action detection problem using two di erent paradigms: \proposal + classi cation" and \end-to-end temporal action detection". In proposal + classi cation approach, the regions which likely to contain human actions, known as proposals, arerst generated from untrimmed videos which are then classi ed into the targeted actions. To this end, we propose two di erent methods to generate action proposals: (1) un-supervised and (2) supervised temporal action proposal methods. In therst method, we propose unsupervised proposal generation method named as Proposals from Motion History Images (PMHI). PMHI discriminates actions from non-action regions by clustering the MHIs into actions and nonaction segments by detecting minima from the energy of MHIs. The strength of PMHI is that it is unsupervised, which alleviates the requirement for any training data. PMHI outperforms the existing proposal methods on the Multi-view Human Action video (MuHAVi)- uncut and Computer Vision and Pattern recognition (CVPR) 2012 Change Detection datasets.PMHI depends upon precise silhouettes extraction which is challenging for realistic videos and for moving cameras. To solve aforementioned problem, we propose a supervised temporal action proposal method named as Temporally Aggregated Bag-of-Discriminant-Words (TAB) which work directly on RGB videos. TAB is based on the observation that there are many overlapping frames in action and background temporal regions of untrimmed videos, which cause di culties in segmenting actions from non-action regions. TAB solve this issue by extracting class-speci c codewords from the action and background videos and extracting the discriminative weights of these codewords based on their ability to discriminate between these two classes. We integrate these discriminative weights with Bag of Word encoding, which we then call Bag-of-Discriminant-Words (BoDW). We sample the untrimmed videos into non-overlapping snippets and temporally aggregate the BoDW representation of multiple snippets into action proposals. We present the e ectiveness of TAB proposal method on two challenging temporal action detection datasets: MSR-II and Thumos14, where it improves upon state-ofthe- art methods. \Proposal + classi cation", requires multiple passes through testing data for these two stages, therefore, it is di cult to use these methods in an end-to-end manner. To solve this problem, we propose an end-to-end temporal action detection method known as Bag of Discriminant Snippets (BoDS). BoDS is based on the observation that multiple actions and the background classes have similar snippets, which cause incorrect classi cation of action regions and imprecise boundaries. We solve this issue bynding the key-snippets from the training data of each class and compute their discriminative power which is used in BoDS encoding. During testing of an untrimmed video, wend the BoDS representation for multiple candidate regions andnd their class label based on a majority voting scheme. We test BoDS on the Thumos14 and ActivityNet datasets and obtain state-of-the-art results.
"کہاں ہیں میرے وہ شریک جن کا تم بڑا دعوی کرتے تھے؟"۔
یعنی وہ بت یا اشخاص ہیں، جن کو تم دنیا میں میری الوہیت میں شریک گردانتے تھے، انھیں مدد کے لئے پکارتے تھے اور ان کے نام کی نذر نیاز دیتے تھے، آج کہاں ہیں؟ کیا وہ تمہاری مدد کر سکتے ہیں اور تمہیں میرے عذاب سے چھڑا سکتے ہیں؟ یہ تقریع وتوبیخ کے طور پر اللہ تعالیٰ ان سے کہے گا، ورنہ وہاں اللہ کے سامنے کس کی مجال ہوگی؟
This Paper analyses the causes of the domestic violence against female beggars and their impacts on their lives. The paper is strictly limited to Taluka Qasimabad, District Hyderabad. The domestic violence against women exits in various forms. However, it is pretty horrible in the form being inflicted against female beggars in Qasimabad. The scores of causes for the domestic violence against female beggars range from rising poverty to population explosions with deep physical and mental impacts on the sufferers. For data collection qualitative research through Focused Group discussion method has been used through snow ball technique. 10 Cases of female beggars have been opted for the FGDs out of which two participated with their social backgrounds. The study concludes the pathetic condition of the female beggars due to multiple factors and their serious implications both physically as well as mentally. The study recommends prompt action from the Government as well as social scientist to go deep further in the social issues such as violence against the female beggars. Simultaneously there is immediate intervention of Government and the concerned department to address these issues on emergent basis as the number is likely to reach an alarming level.
In the present work cyclic voltammetric and electron spin resonance (ESR) spectroscopic investigations of fifteen quinones have been carried out. Quinones belong to a class of organic compounds, which find potential applications in biology and chemistry. Five compounds from each of the three series of quinones, namely benzoquinones (BQs), naphthoquinones (NQs) and anthraquinones (AQs), were selected for the present study. Systematic cyclic voltammetric measurements were made on all compounds in solvents dichloromethane (DCM), acetonitrile (AN) and propylene carbonate (PC) at 25 o C. Analysis of the voltammograms provided fundamental electrochemical parameters (redox potentials, peak separation, peak currents, half-wave potential, peak width) which helped in the interpretation of role of solvents, role of structure and the effect of substituents. The redox behaviour of the compounds was examined first within a series and then a comparative analysis of the three series was made. It was found that substituents affect strongly the redox bahaviour of the compounds. The quinones with electron withdrawing groups were easily reduced than those with electron releasing groups. Among the three series the ease of reduction followed the order BQs>NQs>AQs. 2-Hydroxy-1, 4-naphthoquinone behaved differently due to self protonation. The heterogeneous electron transfer rate constants (k o ) of the first and the second reduction steps were determined in the three solvents employing Nicholson and Kochi methods. The experimental results were compared with those calculated theoretically from the modified form of the Marcus theory. The solvent reorganization energy (λ o ) in the Marcus equation was calculated using the conventional spherical as well as multisphere models. Experimental rate constants in acetonitrile from the Kochi’s method for benzoquinones, naphthoquinones and anthraquinones were found in the range 1.65 x 10 -3 – 8.47 x 10 -3 cm s -1 , 7.09 x 10 -3 – 11.16 x 10 -3 cm s -1 , and 0.19 x 10 -3 – 7.10 x 10 -3 cm s -1 respectively. The experimentally determined rate constants for quinones were found in close agreement with the theoretically calculated rate constants from the Kochi’s method. The effect of medium on electron transfer rates was rationalized in terms of solvent properties such as polarity, viscosity, density relaxation time. An increase in the solvent polarity and a decrease in viscosity favoured the heterogeneous iiielectron transfer rate. Electron transfer rates were found inversely proportional to solvent longitudinal relaxation time. Electrochemical behaviour of quinones was also investigated in the presence of tert-butanol, 2-propanol, ethanol and methanol (monoalcohols), ethylene glycol (a diol) and glycerol (a triol) as proton donors. The quinone-alcohol interaction was analyzed from changes observed in the shape and peak position of the redox waves in voltammograms as a result of change in concentration of the added alcohol. An estimate of the strength of the quinone-alcohol interaction which is hydrogen bonding in nature was obtained by calculating the thermodynamic association constants and the number of alcohol molecules attached to anion or dianion of quinones. The interaction with dianion was found much stronger than with the anion. The strength of the hydrogen bond depended upon the basicity of the quinone and acidity of the alcohol. The presence of α- hydrogens in the quinone structure strengthened the interaction. A comparison of the results for monoalcohols, diol and triol shows that the polyalcohols formed stronger hydrogen bonds. The strength of interaction increased with the increase in the number of OH groups. Homogeneous electron transfer rate constants of quinones in acetonitrile were determined from ESR spectroscopic measurements at 298K. The anion radical of the quinone was generated in situ in a locally fabricated esr-electrochemical cell and the hyperfine spectrum recorded. The hyperfine coupling constants and the line widths were determined from the experimental and simulated ESR spectra. The electron self exchange rate constant was determined from the concentration depended chemical line broadening produced by the addition of neutral quinone. The rate constants were found in the range of 5.2 x 10 8 - 4.4 x 10 9 M -1 s -1 . The strength of precursor complex is observed in terms of association constants (K A ) calculated from the experimental electron transfer rate constants are in the range 0.028 – 1.023 M -1 . Theoretical values of K A were calculated using Eigen-Fuoss and reaction zone models. The values calculated with the reaction zone model agreed with the experimentally determined values. The ESR results support the observed trends in electrochemical behaviour of quinones.